Your activity: 4 p.v.

Diagnosis and management of glucose-6-phosphate dehydrogenase (G6PD) deficiency

Diagnosis and management of glucose-6-phosphate dehydrogenase (G6PD) deficiency
Author:
Bertil Glader, MD, PhD
Section Editor:
Robert T Means, Jr, MD, MACP
Deputy Editor:
Jennifer S Tirnauer, MD
Literature review current through: Dec 2022. | This topic last updated: Dec 14, 2022.

INTRODUCTION — Glucose-6-phosphate dehydrogenase (G6PD) deficiency is an inherited disorder caused by a genetic defect in the red blood cell (RBC) enzyme G6PD, which generates NADPH and protects RBCs from oxidative injury. G6PD deficiency is the most common enzymatic disorder of RBCs.

The severity of hemolytic anemia varies among individuals with G6PD deficiency, making diagnosis more challenging in some cases. Identification of G6PD deficiency and patient education regarding safe and unsafe medications and foods is critical to preventing future episodes of hemolysis.

This topic review discusses the clinical manifestations, diagnosis, and management of G6PD deficiency. Separate topic reviews discuss the pathogenesis of G6PD deficiency and an overall approach to the patient with unexplained hemolytic anemia.

Pathophysiology and genetics of G6PD deficiency – (See "Genetics and pathophysiology of glucose-6-phosphate dehydrogenase (G6PD) deficiency".)

Diagnostic approach to the child with hemolytic anemia – (See "Overview of hemolytic anemias in children".)

Diagnostic approach to the adult with hemolytic anemia – (See "Diagnosis of hemolytic anemia in adults".)

COVID-19 CONSIDERATIONS — Since G6PD deficiency is common (see 'Epidemiology' below), it is likely that individuals with G6PD deficiency will become infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the virus that causes coronavirus disease 2019 (COVID-19). It is also likely that some individuals with COVID-19 will have undiagnosed G6PD deficiency.

The following considerations may apply to individuals with known G6PD deficiency:

The oxidant stress of SARS-CoV-2 infection may lead to hemolysis (or worsening hemolysis) in individuals with G6PD deficiency, similar to other acute infectious illnesses. (See 'Inciting drugs, chemicals, foods, illnesses' below.)

Aside from possible worsening of hemolysis, it is not known whether other aspects of SARS-CoV-2 infection or the inflammatory response to the infection differ in individuals with G6PD deficiency. (See "COVID-19: Clinical features", section on 'Severity of symptomatic infection'.)

Medications that may be under study or used for treatment of COVID-19 are discussed below. (See 'Avoidance of unsafe drugs and chemicals' below.)

It is also possible that the diagnosis of G6PD deficiency may come to medical attention during hospitalization and/or treatment for COVID-19 [1-3]. Individuals with COVID-19 who develop hemolytic anemia should be assessed for G6PD deficiency as part of the comprehensive evaluation. (See "Diagnosis of hemolytic anemia in adults", section on 'Post-diagnostic testing to determine the cause'.)

It may also be reasonable to determine G6PD status in individuals who have COVID-19 (mild, moderate, or severe), especially if they have clinical worsening [4].

EPIDEMIOLOGY — G6PD deficiency is the most common enzymatic disorder of red blood cells (RBCs), affecting 400 to 500 million people worldwide [5-9].

G6PD deficiency is global in its distribution (figure 1) [8,9]. It occurs most often in the tropical and subtropical zones of the Eastern Hemisphere (eg, Africa, Europe, Asia), with prevalences of 20 percent or more in some regions [9-11]. The following prevalences have been reported:

Kurdish Jews – 60 to 70 percent [12]

Sardinians – 4 to 35 percent, depending on the village [13]

Nigerians – 22 percent [14]

Thai people – 17 percent [15]

African Americans – 11 to 12 percent [16,17]

Black Brazilians – 8 percent [18]

Greeks – 6 percent [19]

People from South China – approximately 6 percent [20]

People from India – 3 percent [21]

Japanese and Koreans – 0 to 1 percent [22,23]

This geographic distribution is highly correlated with regions in which malaria was once endemic, leading to the hypothesis that G6PD deficiency may have conferred a selective advantage against infection by Plasmodium falciparum, similar to observations with other RBC abnormalities. (See "Protection against malaria in the hemoglobinopathies" and "Protection against malaria by abnormalities in red cell surface antigens and cytoskeletal proteins".)

VARIATION IN DISEASE SEVERITY

Genetics — G6PD deficiency is an X-linked disorder. As a result, males who inherit a G6PD mutation are hemizygous for the defect; all of their RBCs are affected. (See "Inheritance patterns of monogenic disorders (Mendelian and non-Mendelian)", section on 'Sex-linked'.)

Females who inherit a heterozygous G6PD mutation usually do not have severe hemolytic anemia, since half of their RBCs express the normal G6PD allele and half express the abnormal allele. The majority of females who inherit an abnormality in G6PD are unaffected carriers. However, the cells that express the abnormal allele are as vulnerable to hemolysis as the enzyme-deficient RBCs in males. (See "Genetics and pathophysiology of glucose-6-phosphate dehydrogenase (G6PD) deficiency", section on 'Pathophysiology of G6PD deficiency'.)

The presence of anemia will vary depending on the severity of deficiency in the affected cells and whether there is skewed X-inactivation (lyonization) that results in a greater expression from the abnormal allele in a large percentage of RBCs [24].

Homozygosity or compound heterozygosity for an abnormal G6PD gene has been reported in as many as 1 percent of American women [24-26]. These females are as severely affected as males.

Additional details about the genetics of G6PD deficiency are presented separately. (See "Genetics and pathophysiology of glucose-6-phosphate dehydrogenase (G6PD) deficiency", section on 'Genetics of G6PD'.)

Variant classification — Over 200 G6PD variants have been described [9]. These have been classified by the World Health Organization according to the magnitude of the enzyme deficiency and the severity of hemolysis [27,28]. This classification gives some approximation of the magnitude of hemolysis an individual might incur in the setting of an oxidative stress. Only class I, II, and III are of clinical significance.

Class I – Class I variants have severe enzyme deficiency (eg, <10 percent of normal) associated with chronic hemolytic anemia.

Class II – Class II variants also have severe enzyme deficiency (<10 percent of normal), but there is usually only intermittent hemolysis, typically on exposure to oxidant stress such as fava bean exposure or ingestion of certain drugs. G6PD Mediterranean is the classic example.

Class III – Class III variants have moderate enzyme deficiency (10 to 60 percent of normal) with intermittent hemolysis, typically associated with significant oxidant stress. G6PD A- (the most common variant in individuals of African ancestry) is the classic example.

Class IV – Class IV variants have no enzyme deficiency or hemolysis. The wild-type (normal) enzyme is considered a class IV variant, as are numerous other genetic changes that do not alter levels of the enzyme. These variants are of no clinical significance.

Class V – Class V variants have increased enzyme activity (more than twice normal). These are typically uncovered during testing for G6PD deficiency. They are of no clinical significance.

Some variants are seen more commonly in certain populations:

G6PD Mediterranean – The G6PD Mediterranean variant (563C>T) is the most common abnormal variant in individuals from the Mediterranean region and the Middle East [12]. It is a class II variant associated with severe hemolysis. The half-life of this variant is measured in hours. Thus, the majority of circulating RBCs have grossly deficient G6PD enzyme activity and will undergo hemolysis upon exposure to an oxidant stress. However, in the absence of oxidant stress, hemolysis typically does not occur and there is no anemia or reticulocytosis.

G6PD A- – The G6PD A- variant (202G>A/376A>G) is the most common variant in individuals of African ancestry. It is a class III variant associated with mild to moderate hemolysis as well as sensitivity to the antimalarial drug primaquine. The enzymatic activity of this variant is normal in reticulocytes, but it declines more rapidly than the normal enzyme (half-life, 13 days, compared with 62 days for the normal enzyme) [29,30]. As a result, only the oldest RBCs undergo hemolysis upon exposure to oxidant stress.

Variants in individuals with ancestry from countries in Asia – Several different G6PD variants occur in individuals with Asian ancestry [31,32].

In China, three major variants are recognized. The most common is G6PD Canton (1376G>T), which is usually reported to be a class II variant, although sometimes it is considered to be in class III. Another common variant is G6PD Kaiping (1388G>A), which is usually classified as a class III variant although occasionally considered to be in class II. A third variant is G6PD Gaohe (95A>G), which is almost always considered to be a class II variant. These three variants account for over 70 percent of G6PD deficiency cases in China.

The most common variant in Southeast Asia is G6PD Mahidol (487G>A), a class III variant.

Of interest, although India borders China, none of the Chinese G6PD variants are found in India, where the most common type is G6PD Mediterranean (563C>T).

The genetics and pathophysiology of G6PD deficiency is discussed in more detail separately. (See "Genetics and pathophysiology of glucose-6-phosphate dehydrogenase (G6PD) deficiency", section on 'Classification of G6PD variants'.)

CLINICAL MANIFESTATIONS — The clinical expression of G6PD deficiency encompasses a spectrum of disease severity. The severity of disease and the likelihood of developing neonatal jaundice or chronic hemolysis, and the magnitude of hemolysis when hemolytic episodes occur, depends on the degree of the enzyme deficiency, which in turn is determined by the characteristics of the G6PD variant. (See 'Variant classification' above.)

The majority of individuals are asymptomatic and do not have hemolysis in the steady state [9]. They have neither anemia, evidence of increased red blood cell (RBC) destruction, nor alteration in blood morphology, although a modest shortening of RBC survival can be demonstrated by isotopic techniques [33,34]. This includes almost all individuals with the most prevalent G6PD variants, G6PD A- and G6PD Mediterranean. However, episodes of acute hemolysis with hemolytic anemia may be triggered by medications, certain foods (especially fava beans), and acute illnesses, especially infections. (See 'Acute hemolytic anemia' below.)

The rare individuals with severe disease (class I variants) usually have chronic hemolysis. (See 'Congenital nonspherocytic hemolytic anemia and chronic hemolysis' below.)

In many cases, women who have inherited one abnormal G6PD allele are unaffected carriers. However, as noted above, those with skewed lyonization may have clinical disease. Women who are homozygous or compound heterozygous for an abnormal G6PD gene will have a phenotype similar to men. (See 'Epidemiology' above.)

Acute hemolytic anemia — Some individuals with G6PD deficiency have episodes of acute hemolysis in the setting of oxidant injury from medications, acute illnesses, and certain foods such as fava beans (table 1). Once patients are diagnosed and are able to reduce oxidant stress exposures through medication avoidance, the frequency of hemolysis may decline dramatically. Episodes of acute hemolysis are more common in individuals with G6PD Mediterranean, which has a half-life measured in hours, than with G6PD A-, which has a half-life measured in days.

G6PD-deficiency-associated hemolytic anemia can falsely lower the HbA1C level due to increased RBC turnover, as with other hemolytic anemias, but G6PD deficiency does not alter the HbA1C by mechanisms other than hemolysis. (See "Measurements of glycemia in diabetes mellitus", section on 'Unexpected or discordant values'.)

Typical presentation and blood smear findings — The typical episode of hemolysis is illustrated by the course of an acute hemolytic episode following the administration of primaquine to an individual with G6PD A-, the variant most commonly seen in individuals of African ancestry [35].

At two to four days after drug ingestion, there is the sudden onset of jaundice, pallor, and dark urine, with an abrupt fall in the hemoglobin concentration by 3 to 4 g/dL. There may be abdominal pain and/or back pain.

Hemolysis may be mild and self-limiting in some individuals, and severe and life-threatening in others [36]. A series of 94 individuals from the Brazilian Amazon with the African G6PD variant (some heterozygotes and some homozygotes) who were treated with primaquine for malaria found a high frequency of severe anemia (59 percent), hemoglobinuria (48 percent), and acute kidney injury (28 percent) [37]. Half required blood transfusion, four were admitted to the intensive care unit, and one died. An editorial commented that point-of-care rapid diagnostic testing for G6PD deficiency could be especially helpful in settings such as this. (See 'Point of care tests under investigation' below.)

Findings on the peripheral blood smear reveals microspherocytes, eccentrocytes or "bite" cells, and "blister cells" with hemoglobin puddled to one side (picture 1). Special stains can document the production of Heinz bodies, which are collections of denatured globin chains often attached to the RBC membrane. Hemolysis is both extravascular and intravascular. (See "Unstable hemoglobin variants", section on 'Hemoglobin precipitation and Heinz body formation'.)

The anemia induces an appropriate stimulation of erythropoiesis, with an increase in reticulocytes that is apparent within five days and is maximal at 7 to 10 days after the onset of hemolysis. These reticulocytes and younger RBCs have the highest levels of G6PD activity, often sufficient to withstand the oxidative stress of ongoing drug exposure. As a result, the acute hemolytic process ends after approximately one week, with ultimate reversal of the anemia, even with continued drug ingestion.

More severe hemolysis may be seen in individuals with G6PD Mediterranean, the variant most commonly seen in individuals from Mediterranean countries and the Middle East, as well as in other class II variants (see 'Variant classification' above). The anemia is more severe because a larger population of circulating RBCs is vulnerable to hemolysis, since the half-life of G6PD Mediterranean is shorter and fewer RBCs have sufficient G6PD activity to prevent oxidant injury. Hemolysis in these individuals may continue well after the drug is discontinued [38,39].

The clinical presentation of G6PD-deficiency in individuals with Asian ancestry is variable, depending whether the defect is a class II or III variant. (See 'Variant classification' above.)

Inciting drugs, chemicals, foods, illnesses — Sources of oxidant injury that may elicit an episode of acute hemolysis in an individual with G6PD include a number of medications and chemicals, as well as several foods and certain acute illnesses, especially infections. In a series of 102 patients with G6PD deficiency that categorized 119 episodes of acute hemolysis, 46 (39 percent) were precipitated by a medication alone, and 73 (61 percent) appeared to be related solely to a concurrent illness [40]. Avoiding these inciting substances is important in management. (See 'Avoidance of unsafe drugs and chemicals' below.)

Medications and chemicals – Medications that can precipitate hemolysis in G6PD-deficient individuals are listed in the table (table 1). Classic examples include the antibiotics primaquine and dapsone and the anti-uricemic drugs rasburicase and pegloticase. Additional chemicals such as henna compounds used in hair dyes and tattoos, aniline dyes, and naphthaline (found in moth balls and lavatory deodorants) may also cause hemolysis [41]. Safe amounts of these drugs or safe G6PD activity levels have not been established, so avoidance is the best approach.

Conflicting information may be found regarding certain medications, such as those that modestly shorten the RBC lifespan. These may appear on some lists as "safe" and others as "unsafe" [42]. One such example is sulfamethoxazole, a component of the commonly used trimethoprim-sulfamethoxazole [42-44]. Sulfamethoxazole has been widely used, and cases of well-documented hemolysis in individuals with G6PD deficiency are uncommon. Clinical judgment should be used in deciding if it is safe for a specific individual; if it was used previously and found to be safe (eg, before the diagnosis was made) it may be reasonable to treat it as safe for that individual. A similar approach may be used for other sulfa-containing medications. Other drugs may have initially been labeled as "unsafe" when in fact hemolysis was caused by an infection that the drug was administered to treat (eg, aspirin).

Chloroquine and hydroxychloroquine are listed in some tables of unsafe drugs, but many experts believe these are probably safe when used in standard doses [42,45,46]. Off-label use of these drugs has occurred in patients with COVID-19, and a few scattered case reports suggest that hydroxychloroquine causes acute hemolysis [3,47]. However the causal relationship between hydroxychloroquine and hemolysis is unclear, since it is difficult to separate out the role of infection itself as a hemolytic trigger [48].

The sexual enhancement drug "RUSH," which may contain amyl nitrite or isobutyl nitrite, has been reported to cause hemolysis in individuals with G6PD deficiency [49]. This drug can also cause methemoglobinemia. (See "Methemoglobinemia", section on 'Acquired causes'.)

The common denominator of drugs that can precipitate hemolysis is their interaction with hemoglobin and oxygen, leading to the formation of H2O2 and other oxidizing radicals within RBCs [10,11,42,50]. As these oxidants accumulate within enzyme-deficient RBCs, glutathione levels are reduced, while hemoglobin and other proteins are oxidized, thereby leading to cell lysis. (See "Genetics and pathophysiology of glucose-6-phosphate dehydrogenase (G6PD) deficiency", section on 'Pathophysiology of G6PD deficiency' and "Drug-induced hemolytic anemia", section on 'Oxidant injury'.)

Foods – Certain foods can also trigger episodes of hemolysis in individuals with G6PD deficiency. Ingestion of fava beans, the classic example, is probably the most common cause of hemolysis in individuals with G6PD deficiency [9].

Fava beans can be eaten raw, cooked, or dried (picture 2). They contain high concentrations of two beta-glucosides, vicine and convicine. After fava bean ingestion, vicine and convicine undergo hydrolysis by glucosidases present in both the beans and the gastrointestinal tract. This releases the respective aglycones divicine and isouramil, both of which are highly reactive redox compounds that generate oxygen free radicals and hydrogen peroxide (H2O2).

Acute intravascular hemolysis upon ingestion of fava beans, referred to as favism, occurs most commonly in male children between the ages of one and five years. However, female children and adults of either sex can also be affected [9]. The ratio of fava beans consumed to body weight may account in large part for the observation that favism attacks are much more common and more severe in children than adults [9].

Symptoms of an attack begin within 5 to 24 hours after ingestion of the food and include headache, nausea, back pain, chills, and fever, and are followed by hemoglobinuria and jaundice [51,52]. The fall in hemoglobin concentration is acute, often severe, and, in the absence of transfusion, can be fatal.

Other foods such as bitter melon have also been implicated; these foods are listed below. (See 'Dietary restrictions' below.)

Medical illnesses – Infection is the typical illness that causes hemolysis in G6PD-deficient individuals, and it is likely to be the most common inciting factor for hemolytic anemia once the individual is aware of the diagnosis and avoids oxidant medications. Hemolysis can occur with a variety of organisms (eg, viral, bacterial, rickettsial) and sites of infection (eg, pneumonia, hepatitis). Hemolytic anemia associated with infections can range from mild and self-limited to severe enough to cause acute renal failure [40,53-60]. In a series of patients from a classic report in 1966, pneumonia was the most common inciting infection [40]. In the setting of viral hepatitis, the combination of an increased bilirubin load from hemolysis and a damaged liver unable to process bilirubin as well as normal results in an exaggerated elevation in the serum bilirubin concentration.

The factors responsible for accelerated destruction of G6PD-deficient RBCs during infection are not known. One possible explanation is that the cells are damaged by oxidants generated by phagocytic macrophages [61].

Diabetic ketoacidosis has also been reported to precipitate hemolysis in individuals with G6PD deficiency, although one study of patients with the G6PD Mediterranean variant found no such correlation [40,62,63]. Both acidosis and hyperglycemia are potential precipitating factors, and correction of the abnormalities has been associated with reversal of the hemolytic process [64]. In some diabetic patients, occult infection may be a common trigger for both acute hemolysis and ketoacidosis.

Neonatal jaundice — Anemia and jaundice are often first noted in the newborn period in individuals with severe G6PD deficiency (Class I variants). The degree of jaundice is quite variable; in severe cases, there is a risk of bilirubin-induced neurologic dysfunction and kernicterus (permanent neurologic damage) if the patient is not treated aggressively [65]. (See "Risk factors, clinical manifestations, and neurologic complications of neonatal unconjugated hyperbilirubinemia", section on 'Clinical manifestations'.)

In neonates with Class II or III G6PD-deficiency, jaundice is rarely present at birth; the peak of onset is two to three days after birth [66]. Jaundice is more prominent than anemia, which is rarely severe. Jaundice can be seen in neonates with G6PD Mediterranean, Asian, and African American variants. Monitoring of jaundice and serum bilirubin levels in infants known to be G6PD-deficient is critical [67,68].

The risk of neonatal hyperbilirubinemia associated with G6PD deficiency was illustrated in a 2015 meta-analysis of cohort studies that included 21,585 neonates, 877 of whom had hyperbilirubinemia [69]. The relative risk of hyperbilirubinemia in neonates with G6PD deficiency was 3.92 (95% CI 2.13-7.20). Data from the USA Kernicterus Registry from 1992 to 2004, which were not included in the meta-analysis, indicate that over 30 percent of kernicterus cases are associated with G6PD deficiency [65]. Thus, routine testing for G6PD deficiency is performed in many neonates with hyperbilirubinemia and/or those with less dramatic bilirubin elevations who are of Mediterranean, Nigerian, or East-Asian ancestry. (See 'Diagnostic evaluation' below and "Screening for hyperbilirubinemia in term and late preterm newborn infants", section on 'Subsequent evaluation and management'.)

Observations related to specific populations include the following:

Neonates with the rare class I variants are at greatest risk of neonatal jaundice; however, most infants with hyperbilirubinemia due to G6PD deficiency have more common variants and come from the Mediterranean region or Asia [70-73]. In one series of 43 cases from Italy, for example, 39 had G6PD Mediterranean, one had G6PD A-, and three had other variants [70]. Among affected Chinese children, most cases are associated with G6PD Canton [72].

The risk of neonatal hyperbilirubinemia is less in African-Americans than that in Africans and Jamaicans, despite both groups generally having the same G6PD A- variant [74-76]. Untreated hyperbilirubinemia in Black African and Black Jamaican infants frequently leads to kernicterus with severe neurologic injury or death [75,77]. Likewise, the risk of neonatal hyperbilirubinemia is less in infants of Greek ancestry who are born in Greece than in infants of Greek ancestry born in Australia [78]. These differences in risk with the same G6PD variant may be related to local customs and differences in oxidant exposure. (See 'Inciting drugs, chemicals, foods, illnesses' above.)

The cause of neonatal hyperbilirubinemia in G6PD-deficient infants is not clear [79,80]. It has been presumed that the combination of increased bilirubin production due to accelerated breakdown of RBCs and the immaturity of the liver is responsible [75,81,82]. Indirect evidence such as a lower incidence of neonatal hyperbilirubinemia in immigrants to the United States supports the importance of local environmental variables, although often there is no obvious oxidant exposure [78,83]. Possible exposures may include maternal ingestion of oxidant foods, herbs used in traditional Chinese medicine, and clothing impregnated with naphthalene [84,85]. Some neonates with G6PD Mediterranean have a partial defect in bilirubin glucuronide conjugation similar to that seen in Gilbert's disease [86]. In support of this hypothesis are the observations that carboxyhemoglobin production, a marker of hemolysis or RBC breakdown, is the same in G6PD Mediterranean deficient neonates with and without hyperbilirubinemia [79]. The relative importance of this Gilbert variant is supported by the observed lack of anemia and hemolysis in most jaundiced G6PD-deficient neonates [87]. (See "Etiology and pathogenesis of neonatal unconjugated hyperbilirubinemia".)

In the United States, there is concern that changes in health care delivery with early discharge of newborn infants may increase the risk. One report described four newborn infants with G6PD deficiency (three African-American, and one mixed Peruvian/Chinese) who developed kernicterus following early hospital discharge, even though there was adherence to the early neonatal discharge guidelines of the American Academy of Pediatrics and the American College of Obstetricians and Gynecologists [88].

Congenital nonspherocytic hemolytic anemia and chronic hemolysis — Chronic hemolysis is not characteristic of most individuals with G6PD deficiency, but some with severe deficiency (eg, activity <10 percent at baseline) can have chronic hemolysis with or without chronic anemia. Variants that produce chronic hemolytic anemia are referred to as class I variants (see 'Variant classification' above). These individuals have such severe G6PD deficiency that they may have hemolysis even in the absence of oxidant injury from medications or illnesses [89-93].

These individuals may also be referred to as having congenital nonspherocytic hemolytic anemia. The term nonspherocytic is somewhat of a misnomer, since these individuals may have spherocytes on the peripheral blood smear. However, this term is useful in distinguishing individuals with G6PD deficiency, in whom spherocytes are relatively infrequent at baseline, from those with hereditary spherocytosis, in whom spherocytes are abundant. (See 'Differential diagnosis' below.)

Most individuals with chronic hemolysis have mild to moderate anemia (hemoglobin 8 to 10 g/dL) with a reticulocyte count of 10 to 15 percent. Pallor is uncommon, scleral icterus is intermittent, and splenomegaly is rare. Hemolysis can be exaggerated by exposure to drugs or chemicals with oxidant potential or exposure to fava beans [93]. Some drugs with relatively mild oxidant potential that are safe in patients with class II or class III G6PD variants may increase hemolysis in patients with class I variants.

The typically mild degree of anemia reflects the ability of increased erythropoiesis to compensate for the hemolysis. Thus, as with other chronic hemolytic anemias, the anemia may be worsened by diminished erythropoietic capacity due to infection or to parvovirus-induced aplastic crises. Such a crisis may be the event that first leads to examination of the blood and establishment of diagnosis of G6PD deficiency. (See "Clinical manifestations and diagnosis of parvovirus B19 infection".)

Neutrophil dysfunction — G6PD is used by other cells besides RBCs to reduce oxidant injury. Rarely, individuals with severe G6PD deficiency (eg, <20 percent activity at baseline) may have neutrophil dysfunction due to an impaired respiratory burst, with impaired bactericidal activity and recurrent infections with catalase-positive organisms [94]. Although this has been reported, in our clinical experience patients do not appear to be more susceptible to infections.

This subject is discussed in more detail separately. (See "Myeloperoxidase deficiency and other enzymatic WBC defects causing immunodeficiency", section on 'Glucose-6-phosphate dehydrogenase deficiency'.)

DIAGNOSTIC EVALUATION

Indications for evaluation — Testing for G6PD deficiency may be appropriate in the following settings:

Evaluation of neonatal jaundice or unexplained hemolytic anemia. (See 'Patients being evaluated for the cause of neonatal jaundice or hemolysis' below.)

Asymptomatic individuals at high risk of G6PD deficiency prior to administration of certain medications. (See 'Patients at risk for G6PD deficiency who require treatment with an oxidant medication' below.)

Certain other populations (eg, certain newborn screening settings or asymptomatic family members of affected individuals). (See 'Role of population screening' below and 'Genetic counseling and prenatal testing' below.)

Testing may be performed using an initial screening test followed by a confirmatory test, or using the confirmatory test initially, depending on available resources and institutional guidelines. (See 'Screening tests' below and 'Confirmatory tests' below.)

Patients being evaluated for the cause of neonatal jaundice or hemolysis — Testing for G6PD deficiency is appropriate in infants with unexplained neonatal jaundice and in an individual of any age with unexplained, direct antiglobulin (Coombs) test (DAT)-negative hemolytic anemia, especially those from families with a history of inherited anemia and those from populations most likely to be affected (eg, African, Southern European, Middle Eastern, Chinese, or Southeast Asian heritage). (See 'Epidemiology' above.)

In those with a strong suspicion for G6PD deficiency, this testing may be done early in the evaluation. In others, it may be done following negative testing for more likely causes of unexplained anemia or hemolytic anemia. (See "Approach to the child with anemia" and "Diagnostic approach to anemia in adults" and "Diagnosis of hemolytic anemia in adults".)

Patients at risk for G6PD deficiency who require treatment with an oxidant medication — Testing for G6PD deficiency is also appropriate for individuals who require treatment with oxidant drugs including dabrafenib, dapsone, chlorpropamide, glipizide, glyburide, methylene blue, pegloticase, primaquine, quinine, rasburicase, and others. A common example is a patient who requires presumptive anti-relapse therapy with primaquine to eradicate the liver stages of Plasmodium vivax or P. ovale. (See "Non-falciparum malaria: P. vivax, P. ovale, and P. malariae", section on 'Preventing relapse' and "Prevention of malaria infection in travelers".)

In most cases, it is prudent to screen for G6PD deficiency if the clinician needs to use a drug that could potentially cause oxidant injury (eg, patients with HIV infection who require dapsone) or treatment with rasburicase, which can cause hemolysis in individuals with G6PD deficiency. However, there may be cases in which testing can reasonably be omitted, such as treatment with a sulfonylurea or nitrofurantoin, which only cause mild hemolysis, or if the individual has already received the particular drug in the past (or is currently receiving the drug) without appreciable symptoms of hemolysis.

It is possible that newly available drugs may cause hemolysis in individuals with G6PD deficiency; however, a single case report showing an association is generally not sufficient to infer causation.

Testing for G6PD deficiency prior to administration of medicines that can produce oxidant injury is consistent with information provided in drug labeling by the US Food and Drug Administration (FDA), as summarized on a website on pharmacogenetics hosted by the FDA [95]. Specific drug label information should be consulted to determine whether patients with G6PD deficiency should consider the drug to be contraindicated or whether the drug can be administered with increased monitoring.

Timing of G6PD assay — An important aspect of G6PD diagnosis deficiency is that testing is based on direct measurements of G6PD activity in a population of RBCs. In the setting of an acute hemolytic episode, the RBCs with the most severely reduced G6PD activity will have hemolyzed, and thus their G6PD activity will not be measured in the assay. This situation can produce false-negative results in some patients who are tested in the midst of a severe hemolytic episode.

False-negative results are most likely to occur in individuals who have populations of RBCs that are severely G6PD-deficient and RBCs that are not severely deficient, such as individuals of African ancestry, in whom G6PD activity declines gradually as RBCs age, and women, most of whom are heterozygous and have a mixture of normal and G6PD-deficient RBCs. False-negative results are also most likely to be seen during the period of initial reticulocytosis, when there is the highest proportion of reticulocytes, which typically have normal G6PD activity.

Thus, if initial testing is negative and a suspicion for G6PD deficiency remains, testing should be repeated approximately three months after the hemolytic episode has resolved (ie, the typical time it takes for a new population of circulating RBCs to be produced). During this three-month period, it would be prudent to avoid potential sources of oxidant injury. (See 'Avoidance of unsafe drugs and chemicals' below and 'Dietary restrictions' below.)

Screening tests — Many medical center laboratories perform qualitative screening tests for G6PD deficiency that can provide results in a short period of time (one to two hours). These assays all work by assaying the normal function of the G6PD enzyme, reduction of NADP (nicotinamide adenine dinucleotide phosphate) to NADPH (figure 2), which is the initial step in the hexose monophosphate (HMP) shunt.

In the screening test, adequate G6PD activity with NADPH generation is determined by monitoring a florescent spot under ultraviolet light. For the most part, screening tests are semiquantitative. Thus, if positive, they typically should be followed by a quantitative confirmatory test (see 'Confirmatory tests' below). However, under conditions where it is urgent to know G6PD status (eg, administration of rasburicase prior to initiating chemotherapy), results from the screening test can be used for guidance. It has been suggested that any institution that treats patients with acute leukemia or lymphoma who may require emergency screening for G6PD deficiency should have an on-site test available [50].

Confirmatory tests — Confirmatory testing is performed for individuals with a positive screening test (or in some cases as the initial test) depending on available resources, cost, and institutional practices. Quantitative G6PD assays are available at several reference laboratories (Mayo Clinic Laboratories, Arup Laboratories, Stanford RBC Special Studies Laboratory) [96-98]. Turnaround time for results from outside laboratories is often at least seven days.

The quantitative tests assay NADPH formation quantitatively. They are performed by adding a measured amount of RBC hemolysate to an assay mixture that contains substrate (glucose-6-phosphate) and a cofactor (NADP); the rate of NADPH generation is measured spectrophotometrically (absorbance at a wavelength of 340 nanometers) [99,100].

Results are expressed as units of enzyme activity per gram of hemoglobin. Normal ranges may differ depending on the methodology used and the assay temperature.

Typical normal range at 25°C – 5.5 to 8.8 units/gram of hemoglobin

Typical normal range at 37°C – 8.0 to 13.5 units/gram of hemoglobin

However, a "safe" level of G6PD activity has not been established, and any individual who has had hemolysis related to G6PD deficiency should avoid subsequent exposures regardless of their absolute activity level. (See 'Avoidance of unsafe drugs and chemicals' below.)

Levels of G6PD are higher in the newborn than they are in the adult [26,101]. When higher-than-normal levels are seen in older patients, this almost invariably reflects the presence of a young RBC population with reticulocytosis. Rarely, high activity G6PD variants have been reported, but we do not evaluate for these unless there are extenuating circumstances and the possibility of reticulocytosis has been eliminated. (See 'Variant classification' above.)

Confirmatory testing using molecular/genetic/DNA methods is also available, although this approach is not used routinely. Testing for pathogenic G6PD variants is not particularly useful in the assessment of G6PD-deficient individuals of African or Mediterranean background. However, in Chinese patients where the severity of hemolysis can be variable, molecular studies may be helpful in predicting G6PD class, and thereby assessing potential hemolytic risk. Molecular testing may also be appropriate when it is necessary to identify a heterozygous female with borderline enzyme activity. There are several academic/commercial labs available for gene studies, which can be found on the Genetic Testing Registry website.

Point of care tests under investigation — The need for point-of-care tests to be used on-site (eg, prior to administration of antimalarial drugs) has been emphasized by various groups [102]. Available tests that can be performed on a fingerstick and scored visually (ie, that do not require additional equipment to get the result) have been reviewed [103]. Quantitative point-of-care enzyme testing in potentially affected infants is being assessed as a means of risk reduction for neonatal kernicterus [104]. (See "Risk factors, clinical manifestations, and neurologic complications of neonatal unconjugated hyperbilirubinemia", section on 'Chronic bilirubin encephalopathy (kernicterus)'.)

Role of population screening — The question of whether testing for G6PD deficiency should be included in newborn screening programs worldwide has been raised [105]. This screening has not been widely implemented; however, routine newborn screening is done in some populations with a high incidence of G6PD deficiency. (See "Newborn screening".)

DIFFERENTIAL DIAGNOSIS — The differential diagnosis of G6PD deficiency includes other causes of hemolytic anemia and other causes of neonatal jaundice:

Inherited hemolytic anemias – Other inherited hemolytic anemias include other enzyme deficiencies (eg, pyruvate kinase [PK]), hemoglobinopathies (eg, thalassemia, sickle cell disease), and membrane/cytoskeletal defects (eg, hereditary spherocytosis [HS]). Like G6PD deficiency, these can present with varying degrees of direct antiglobulin (Coombs)-negative hemolysis (with increased reticulocytes, decreased haptoglobin, increased lactate dehydrogenase [LDH], and anemia), and like G6PD deficiency, these may not be diagnosed until adulthood (or may be misclassified), especially if hemolysis is mild. Like many of these inherited disorders, G6PD deficiency is seen at greater frequency in populations from regions where malaria was once endemic, likely due to a protective effect. Unlike these other inherited hemolytic anemias, G6PD deficiency is characterized by low G6PD enzyme activity at baseline, and testing such as hemoglobin analysis, osmotic fragility, and band 3 flow cytometry will be normal in individuals with G6PD deficiency. (See "Overview of hemolytic anemias in children", section on 'Intrinsic hemolytic anemias' and "Diagnosis of hemolytic anemia in adults".)

Acquired hemolytic anemias – Acquired hemolytic anemias include a number of immune and non-immune causes of hemolysis. Like G6PD deficiency, in some cases an acute medical illness or exposure to a drug may precede hemolysis. Like G6PD deficiency, these can present with varying degrees of hemolysis (with increased reticulocytes, decreased haptoglobin, increased LDH, and anemia). Unlike these acquired conditions, G6PD deficiency is characterized by low G6PD enzyme activity at baseline, and testing such as hemoglobin analysis and osmotic fragility will be normal in individuals with G6PD deficiency. (See "Overview of hemolytic anemias in children", section on 'Extrinsic hemolytic anemias' and "Diagnosis of hemolytic anemia in adults" and "Drug-induced hemolytic anemia".)

Conditions causing neonatal hyperbilirubinemia – Neonatal hyperbilirubinemia can be caused by a number of conditions associated with increased bilirubin production (eg, hemolytic disease of the fetus and newborn [HDFN]) or decreased bilirubin clearance (eg, anatomic obstruction, metabolic disorders affecting bilirubin clearance). Like G6PD deficiency, these conditions may be associated with anemia and neonatal jaundice. Unlike G6PD deficiency, these conditions may be more likely to present with hyperbilirubinemia at birth (versus two to three days after birth for G6PD deficiency), and these other conditions are associated with normal G6PD enzyme activity. (See "Etiology and pathogenesis of neonatal unconjugated hyperbilirubinemia".)

MANAGEMENT — The cornerstone of management of G6PD deficiency is the avoidance of oxidative stress to red blood cells (RBCs). This is usually straightforward once the diagnosis is known. However, there may be instances in which an oxidant drug is absolutely required, or cases in which oxidative stress comes from an infection or other acute medical condition that cannot be avoided. In these settings, management depends on the severity of hemolysis and anemia and the patient's age and comorbidities.

Treatment of neonatal jaundice and chronic hemolysis — The management of neonatal jaundice due to G6PD deficiency does not differ from that recommended for neonatal jaundice arising from other causes. Mild cases generally do not require treatment; intermediate cases require phototherapy; and severe cases may require exchange transfusion. (See "Initial management of unconjugated hyperbilirubinemia in term and late preterm newborns".)

For those rare individuals with chronic hemolysis, routine supplementation with folic acid is reasonable. In such cases, a dose of 1 mg daily is adequate. For G6PD-deficient individuals who do not have chronic hemolysis, there is no need for supplemental folic acid.

Treatment of acute hemolytic episodes — Whenever hemolysis occurs in an individual with G6PD deficiency, any inciting agent(s) should be removed as soon as possible [10].

Other interventions may include aggressive hydration for acute intravascular hemolysis or transfusion for severe anemia. (See "Indications and hemoglobin thresholds for red blood cell transfusion in the adult".)

Various treatments directed at the source of oxidant injury or NADPH production have been evaluated and found to be ineffective (eg, xylitol, vitamin E) [10,34,106,107].

Avoidance of unsafe drugs and chemicals — The principle intervention for reducing hemolysis in individuals with G6PD deficiency is avoiding exposure to drugs and chemicals known to trigger hemolysis. There is not universal agreement on which drugs are safe; different sources provide lists that may differ slightly [9,10,42]. A list of commonly implicated drugs and chemicals based on a synthesis of information from the literature is provided in the table (table 1). (See 'Inciting drugs, chemicals, foods, illnesses' above.)

There may be certain settings in which it is especially important to give one of these drugs, and this may be possible in individuals with mild hemolysis (eg, class III variants) (see 'Variant classification' above). As examples:

Primaquine antimalarial prophylaxis has been given to individuals with the G6PD A- variant as long as a low dose is used (15 mg/day or 45 mg once or twice weekly) and the complete blood count (CBC) is monitored closely [108]. The mild anemia that may ensue is corrected by the compensatory increase in reticulocyte production and does not recur unless the dose of the drug is escalated. (See "Antimalarial drugs: An overview", section on 'Primaquine'.)

In other cases, the drug may be life-saving and may need to be administered before the results of G6PD testing are available (eg, administration of rasburicase for tumor lysis syndrome). In these situations, it may be prudent to provide the drug and maintain a high index of suspicion for hemolysis that will facilitate rapid treatment if hemolysis occurs. (See "Tumor lysis syndrome: Prevention and treatment", section on 'Rasburicase'.)

Chloroquine or hydroxychloroquine may be listed in some tables of drugs to avoid in individuals with G6PD deficiency; however, many experts consider them to be probably safe when used at standard doses, as discussed above. (See 'Inciting drugs, chemicals, foods, illnesses' above.)

The potential increased risk of hemolysis in an individual with a severe viral infection such as COVID-19 is unknown. (See 'COVID-19 considerations' above.)

In other cases, an alternative drug may be effective. The use of ascorbic acid (vitamin C) rather than methylene blue to treat methemoglobinemia in individuals with G6PD deficiency is discussed separately. (See "Methemoglobinemia", section on 'Ascorbic acid (vitamin C)'.)

There are no high-quality data to suggest a certain level of G6PD deficiency is "safe" to administer these medications.

Data regarding dietary supplements and herbs are challenging to evaluate. In a systematic review of published reports, no evidence of harms were observed for vitamin C, vitamin E, vitamin K, ginkgo biloba, or alpha-lipoic acid [109]. We neither prescribe nor proscribe any of these supplements for our G6PD-deficient patients. Just as for any questionable food, we ask our patients and their families to be observant of any changes suggestive of increased hemolysis (change in stamina, scleral icterus, dark [cola-colored] urine) associated with the use of supplements. There are no data to evaluate whether high doses of these supplements or vitamins are a risk factor for hemolysis.

Dietary restrictions — It has also been known since antiquity that ingestion of fava beans can cause acute hemolytic anemia in some individuals. In 1958, it was recognized that all individuals with favism had G6PD deficiency [110,111]. Thus, individuals with G6PD deficiency should avoid ingestion of fava beans, also referred to as "broad beans," which can cause hemolysis in some but not all affected individuals [52,112]. However, unlike certain medications that induce hemolysis in all individuals with G6PD deficiency, sensitivity to the fava bean is more variable.

The G6PD variant most commonly implicated in favism is G6PD Mediterranean and G6PD Canton. Thus, favism occurs most often in people from Italy, Greece, North Africa, the Middle East, and Asia [10]. Africans and African-Americans with G6PD deficiency are much less susceptible, although there are very rare cases of favism associated with the African variant, G6PD A- [113]. In addition, the response to the bean by the same individual at different times may not be consistent [114]. Other genetic factors, perhaps related to the hepatic metabolism of potentially oxidant compounds within the fava bean, may play a role in determining the severity of the reaction [114-116]. For reasons that are unknown, favism occurs mostly in children [117].

Favism most often results from the ingestion of fresh (rather than preserved) fava beans (picture 2). Consequently, the peak seasonal incidence of favism in Mediterranean regions coincides with harvesting of the bean during April and May [114]. However, equally severe hemolysis can occur after consuming fried fava beans, a popular Chinese snack (picture 2). Favism also has been reported in nursing infants whose mothers have eaten fava beans.

A question that often comes up relates to the safety of falafel, a common Middle Eastern food. The answer depends on the ingredients. Egyptian falafel is made from fava beans, whereas falafel made elsewhere in the world is usually made from chick peas, which are considered safe for people with G6PD deficiency. However, in some areas, falafel is made from a mixture of fava beans and chick peas. The easiest response to the question is that G6PD-deficient individuals should not consume anything with fava beans. However, since not everyone with G6PD deficiency, particularly adults, is sensitive to fava beans, we advise patients to use cautious observation.

Favism can also occur following ingestion of bitter melon. Also, several other foods such as blueberries are listed on the internet as potentially associated with hemolysis, although the direct relationship is not clear. In our practice, aside from advising people to avoid fava beans, we do not advise dietary restrictions; however, for any questionable food we emphasize to our patients and their families to be observant of any changes suggestive of increased hemolysis. Patients are encouraged to call their physician if any changes are noted.

The mechanism by which fava beans induce hemolysis involves the pyrimidine metabolites divicine and isouramil (aglycones of the glucosides) [118-120]. These compounds act as strong reducing agents, which in the presence of oxygen form an unstable intermediate that oxidizes reduced glutathione. In G6PD-deficient red blood cells with diminished GSH-generating capacity this may have a direct effect on RBC function. In vitro studies have shown that divicine reduces the activity of catalase which, like the glutathione pathway, contributes to hydrogen peroxide removal, and requires NADPH for maintenance of normal activity [120].

Pregnancy — Overall, pregnancy in women who are heterozygous for G6PD deficiency is safe. However, these women should avoid drugs and chemicals known to be unsafe (table 1). Some of these drugs can cross the placenta and put the fetus at risk, and some can gain access to breast milk and put the neonate at risk. (See 'Inciting drugs, chemicals, foods, illnesses' above.)

For the drugs that are considered "probably safe," there are no published data to suggest that risk of hemolysis would be increased in a breastfed infant who had G6PD deficiency. Decisions about the use of these drugs will depend on the individual case and the availability of good alternatives, as discussed in topic reviews on specific conditions. (See "Lactational mastitis", section on 'Treatment'.)

Blood donation — As a general rule, donated blood is not screened for G6PD deficiency, and individuals with G6PD deficiency can donate blood as long as they are otherwise able to donate and do not have anemia. This is because the typical lifespan of transfused G6PD-deficient RBCs is thought to be relatively normal, and it is unlikely for a patient to be transfused with multiple units of G6PD-deficient blood and have clinically significant hemolysis, even in areas of high prevalence [10]. (See "Blood donor screening: Overview of recipient and donor protections".)

One exception may be blood used for exchange transfusion of newborn infants, which poses a theoretical risk if a large enough volume of G6PD-deficient cells is transfused. (See "Red blood cell transfusions in the newborn" and "Red blood cell transfusion in infants and children: Administration and complications".)

Genetic counseling and prenatal testing — G6PD deficiency is an X-linked disorder. (See 'Genetics' above.)

Affected males have a 100 percent chance of transmitting the abnormal gene to their daughters, who will be heterozygous. Affected females have a 50 percent chance of transmitting the defect to their sons and daughters.

In general, males who inherit an abnormal G6PD gene are more likely to have clinically significant disease, and heterozygous females are likely to be unaffected carriers. However, females can have hemolysis if they have skewed lyonization or if they are homozygous or compound heterozygous for an abnormal G6PD gene, which can happen in populations with a high prevalence of G6PD deficiency.

Prenatal testing for G6PD deficiency is not routinely performed. (See "Prenatal care: Initial assessment".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Anemia in adults".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

Basics topic (see "Patient education: Glucose-6-phosphate dehydrogenase deficiency (The Basics)")

SUMMARY AND RECOMMENDATIONS

Disease definition and epidemiology – Glucose-6-phosphate dehydrogenase (G6PD) deficiency is the most common inherited red blood cell (RBC) enzyme abnormality, affecting 400 million people worldwide (figure 1). Individuals of certain ethnic groups are at higher risk, including Kurdish Jews; Black individuals from sub-Saharan Africa or Brazil; African Americans; and people from Thailand, Sardinia, Greece, South China, and India (areas where malaria was once endemic). G6PD deficiency is an X-linked disorder. Males are more likely to be affected, and heterozygous females are typically unaffected carriers, but females who are homozygous, compound heterozygous, or heterozygous with skewed lyonization can have clinically significant hemolysis. (See 'Epidemiology' above.)

Genetics – Clinically significant G6PD variants are classified as class I, II, or III by presence (class I) or absence of chronic hemolysis and the severity of reduction in enzyme activity (<10 percent for class II; 10 to 60 percent for class III). (See 'Variant classification' above.)

The genetic changes and pathophysiology of G6PD deficiency are presented in detail separately. (See "Genetics and pathophysiology of glucose-6-phosphate dehydrogenase (G6PD) deficiency" and "Gene test interpretation: G6PD".)

Clinical features – Clinical manifestations of G6PD deficiency include acute hemolytic anemia, typically induced by medications (table 1), chemicals (eg, henna, naphthaline), foods such as fava beans (picture 2), or illnesses (typically, infections) that cause oxidant injury. Most individuals have only intermittent episodes of hemolysis, but severely affected individuals can have life-threatening neonatal jaundice and/or chronic hemolytic anemia. (See 'Clinical manifestations' above.)

When to suspect G6PD deficiency – Evaluation for G6PD deficiency is appropriate in individuals with unexplained neonatal jaundice or direct antiglobulin (Coombs)-negative hemolytic anemia, and in high-risk individuals prior to treatment with known oxidant medications (table 1). First-degree relatives of affected individuals and certain other populations may benefit from testing. (See 'Indications for evaluation' above.)

Evaluation – Available testing includes semi-quantitative screening tests, some of which are done at the point-of-care, and quantitative tests that report G6PD enzyme activity per gram of hemoglobin. The principle of these assays is generation of NADPH by RBCs (figure 2). False-negative results may occur in some individuals with acute hemolysis because the most severely G6PD-deficient cells have been destroyed; in such cases, testing should be repeated three months after the hemolytic episode has resolved. DNA testing is available but not used routinely. (See 'Timing of G6PD assay' above and 'Screening tests' above and 'Confirmatory tests' above.)

Differential diagnosis – The differential diagnosis of G6PD deficiency includes a number of other inherited and acquired hemolytic anemias and causes of neonatal jaundice. (See 'Differential diagnosis' above.)

Management – Management depends on the severity of the deficiency and the clinical setting. Specific recommendations for neonatal jaundice, acute hemolytic episodes, chronic hemolysis, and avoidance of unsafe medications (table 1) and foods are presented above. Pregnancy is safe, and individuals with G6PD deficiency can donate blood as long as they are not anemic. (See 'Management' above.)

Other resources – A general approach to the evaluation of patients with hemolytic anemia is also presented separately. (See "Overview of hemolytic anemias in children" and "Diagnosis of hemolytic anemia in adults".)

ACKNOWLEDGMENTS

We are saddened by the death of Stanley L Schrier, MD, who passed away in August 2019. The editors at UpToDate gratefully acknowledge Dr. Schrier's role as Section Editor on this topic, his tenure as the founding Editor-in-Chief for UpToDate in Hematology, and his dedicated and longstanding involvement with the UpToDate program.

The UpToDate editorial staff also acknowledges extensive contributions of Donald H Mahoney, Jr, MD to earlier versions of this topic review.

  1. Al-Abdi S, Al-Aamri M. G6PD deficiency in the COVID-19 pandemic: Ghost within Ghost. Hematol Oncol Stem Cell Ther 2021; 14:84.
  2. Vick DJ. Glucose-6-Phosphate Dehydrogenase Deficiency and COVID-19 Infection. Mayo Clin Proc 2020; 95:1803.
  3. Beauverd Y, Adam Y, Assouline B, Samii K. COVID-19 infection and treatment with hydroxychloroquine cause severe haemolysis crisis in a patient with glucose-6-phosphate dehydrogenase deficiency. Eur J Haematol 2020; 105:357.
  4. Jamerson BD, Haryadi TH, Bohannon A. Glucose-6-Phosphate Dehydrogenase Deficiency: An Actionable Risk Factor for Patients with COVID-19? Arch Med Res 2020; 51:743.
  5. Glader B. Hereditary hemolytic anemias due to red blood cell enzyme disorders. In: Wintrobe's Clinical Hematology, 13th edition, Greer JP, Arber D, Glader B, et al (Eds), Wolters Kluwer/Lippincott Williams & Wilkins, Philadelphia 2014. p.728.
  6. Mason PJ, Bautista JM, Gilsanz F. G6PD deficiency: the genotype-phenotype association. Blood Rev 2007; 21:267.
  7. Cappellini MD, Fiorelli G. Glucose-6-phosphate dehydrogenase deficiency. Lancet 2008; 371:64.
  8. Nkhoma ET, Poole C, Vannappagari V, et al. The global prevalence of glucose-6-phosphate dehydrogenase deficiency: a systematic review and meta-analysis. Blood Cells Mol Dis 2009; 42:267.
  9. Luzzatto L, Ally M, Notaro R. Glucose-6-phosphate dehydrogenase deficiency. Blood 2020; 136:1225.
  10. Beutler E. G6PD deficiency. Blood 1994; 84:3613.
  11. Beutler E. Glucose-6-phosphate dehydrogenase deficiency. N Engl J Med 1991; 324:169.
  12. Oppenheim A, Jury CL, Rund D, et al. G6PD Mediterranean accounts for the high prevalence of G6PD deficiency in Kurdish Jews. Hum Genet 1993; 91:293.
  13. Siniscalco, M, et al. Favism and thalassemia in Sardinia and their relationship to malaria. Nature 1961; 190:1179.
  14. Bienzle U. Glucose-6-phosphate dehydrogenase deficiency. Part 1: Tropical Africa. Clin Haematol 1981; 10:785.
  15. Charoenkwan P, Tantiprabha W, Sirichotiyakul S, et al. Prevalence and molecular characterization of glucose-6-phosphate dehydrogenase deficiency in northern Thailand. Southeast Asian J Trop Med Public Health 2014; 45:187.
  16. Heller P, Best WR, Nelson RB, Becktel J. Clinical implications of sickle-cell trait and glucose-6-phosphate dehydrogenase deficiency in hospitalized black male patients. N Engl J Med 1979; 300:1001.
  17. Chinevere TD, Murray CK, Grant E Jr, et al. Prevalence of glucose-6-phosphate dehydrogenase deficiency in U.S. Army personnel. Mil Med 2006; 171:905.
  18. Saldanha PH, Nóbrega FG, Maia JC. Distribution and heredity of erythrocyte G6PD activity and electrophoretic variants among different racial groups at São Paulo, Brazil. J Med Genet 1969; 6:48.
  19. Stamatoyannopoulos G, Panayotopoulos A, Motulsky AG. The distribution of glucose-6-phosphate dehydrogenase deficiency in Greece. Am J Hum Genet 1966; 18:296.
  20. CHAN TK, TODD D, WONG CC. ERYTHROCYTE GLUCOSE-6-PHOSPHATE DEHYDROGENASE DEFICIENCY IN CHINESE. Br Med J 1964; 2:102.
  21. Panich V. Glucose-6-phosphate dehydrogenase deficiency. Part 2. Tropical Asia. Clin Haematol 1981; 10:800.
  22. Nakatsuji T, Miwa S. Incidence and characteristics of glucose-6-phosphate dehydrogenase variants in Japan. Hum Genet 1979; 51:297.
  23. Goo YK, Ji SY, Shin HI, et al. First evaluation of glucose-6-phosphate dehydrogenase (G6PD) deficiency in vivax malaria endemic regions in the Republic of Korea. PLoS One 2014; 9:e97390.
  24. Hsia YE, Miyakawa F, Baltazar J, et al. Frequency of glucose-6-phosphate dehydrogenase (G6PD) mutations in Chinese, Filipinos, and Laotians from Hawaii. Hum Genet 1993; 92:470.
  25. TARLOV AR, BREWER GJ, CARSON PE, ALVING AS. Primaquine sensitivity. Glucose-6-phosphate dehydrogenase deficiency: an inborn error of metabolism of medical and biological significance. Arch Intern Med 1962; 109:209.
  26. Algur N, Avraham I, Hammerman C, Kaplan M. Quantitative neonatal glucose-6-phosphate dehydrogenase screening: distribution, reference values, and classification by phenotype. J Pediatr 2012; 161:197.
  27. Beutler E. The molecular biology of enzymes of erythrocyte metabolism. In: The Molecular Basis of Blood Disease, Stamatoyannopoulos G, Nienhus AW, Majerus PW, et al. (Eds), WB Saunders, Philadelphia 1993.
  28. Glucose-6-phosphate dehydrogenase deficiency. WHO Working Group. Bull World Health Organ 1989; 67:601.
  29. Piomelli S, Corash LM, Davenport DD, et al. In vivo lability of glucose-6-phosphate dehydrogenase in GdA- and GdMediterranean deficiency. J Clin Invest 1968; 47:940.
  30. Yoshida A, Stamatoyannopoulos G, Motulsky AG. Negro variant of glucose-6-phosphate dehydrogenase deficiency (A-) in man. Science 1967; 155:97.
  31. Jiang W, Yu G, Liu P, et al. Structure and function of glucose-6-phosphate dehydrogenase-deficient variants in Chinese population. Hum Genet 2006; 119:463.
  32. He Y, Zhang Y, Chen X, et al. Glucose-6-phosphate dehydrogenase deficiency in the Han Chinese population: molecular characterization and genotype-phenotype association throughout an activity distribution. Sci Rep 2020; 10:17106.
  33. Brewer GJ, et al. The hemolytic effect of primaquine. XII. Shortened erythrocyte life span in primaquine-sensitive male negroes in the absence of drug administration. J Lab Clin Med 1961; 58:217.
  34. Corash L, Spielberg S, Bartsocas C, et al. Reduced chronic hemolysis during high-dose vitamin E administration in Mediterranean-type glucose-6-phosphate dehydrogenase deficiency. N Engl J Med 1980; 303:416.
  35. DERN RJ, BEUTLER E, ALVING AS. The hemolytic effect of primaquine. II. The natural course of the hemolytic anemia and the mechanism of its self-limited character. J Lab Clin Med 1954; 44:171.
  36. Pamba A, Richardson ND, Carter N, et al. Clinical spectrum and severity of hemolytic anemia in glucose 6-phosphate dehydrogenase-deficient children receiving dapsone. Blood 2012; 120:4123.
  37. Brito-Sousa JD, Santos TC, Avalos S, et al. Clinical Spectrum of Primaquine-induced Hemolysis in Glucose-6-Phosphate Dehydrogenase Deficiency: A 9-Year Hospitalization-based Study From the Brazilian Amazon. Clin Infect Dis 2019; 69:1440.
  38. Pannacciulli I, Salvidio E, Tizianello A, Parravidino G. Hemolytic effects of standard single dosages of primaquine and chloroquine on G-6-PD-deficient caucasians. J Lab Clin Med 1969; 74:653.
  39. George JN, Sears DA, McCurdy PR, Conrad ME. Primaquine sensitivity in Caucasians: hemolytic reactions induced by primaquine in G-6-PD deficient subjects. J Lab Clin Med 1967; 70:80.
  40. Burka ER, Weaver Z 3rd, Marks PA. Clinical spectrum of hemolytic anemia associated with glucose-6-phosphate dehydrogenase deficiency. Ann Intern Med 1966; 64:817.
  41. Raupp P, Hassan JA, Varughese M, Kristiansson B. Henna causes life threatening haemolysis in glucose-6-phosphate dehydrogenase deficiency. Arch Dis Child 2001; 85:411.
  42. Youngster I, Arcavi L, Schechmaster R, et al. Medications and glucose-6-phosphate dehydrogenase deficiency: an evidence-based review. Drug Saf 2010; 33:713.
  43. Markowitz N, Saravolatz LD. Use of trimethoprim-sulfamethoxazole in a glucose-6-phosphate dehydrogenase-deficient population. Rev Infect Dis 1987; 9 Suppl 2:S218.
  44. Beutler E. Disorders of red cells resulting from enzyme abnormalities. In: Williams' Hematology, 7th ed, Lichtman MA, Beutler E, Kipps TJ, et al (Eds), McGraw-Hill, 2006. p.603.
  45. Beutler E. Glucose-6-phosphate dehydrogenase deficiency: a historical perspective. Blood 2008; 111:16.
  46. Schilling WHK, Bancone G, White NJ. No evidence that chloroquine or hydroxychloroquine induce hemolysis in G6PD deficiency. Blood Cells Mol Dis 2020; 85:102484.
  47. Maillart E, Leemans S, Van Noten H, et al. A case report of serious haemolysis in a glucose-6-phosphate dehydrogenase-deficient COVID-19 patient receiving hydroxychloroquine. Infect Dis (Lond) 2020; 52:659.
  48. Afra TP, Vasudevan Nampoothiri R, Razmi T M. Doubtful precipitation of hemolysis by hydroxychloroquine in glucose-6-phosphate dehydrogenase-deficient patient with COVID-19 infection. Eur J Haematol 2020; 105:512.
  49. Beaupre SR, Schiffman FJ. Rush hemolysis. A 'bite-cell' hemolytic anemia associated with volatile liquid nitrite use. Arch Fam Med 1994; 3:545.
  50. Luzzatto L, Seneca E. G6PD deficiency: a classic example of pharmacogenetics with on-going clinical implications. Br J Haematol 2014; 164:469.
  51. Luisada A. Favism. Medicine (Baltimore) 1941; 20:229.
  52. Luzzatto L, Arese P. Favism and Glucose-6-Phosphate Dehydrogenase Deficiency. N Engl J Med 2018; 378:60.
  53. Shannon K, Buchanan GR. Severe hemolytic anemia in black children with glucose-6-phosphate dehydrogenase deficiency. Pediatrics 1982; 70:364.
  54. Whelton A, Donadio JV Jr, Elisberg BL. Acute renal failure complicating rickettsial infections in glucose-6-phosphate dehydrogenase-deficient individuals. Ann Intern Med 1968; 69:323.
  55. Phillips SM, Silvers NP. Glucose-6 phosphate dehydrogenase deficiency, infectious hepatitis, acute hemolysis, and renal failure. Ann Intern Med 1969; 70:99.
  56. Chan TK, Chesterman CN, McFadzean AJ, Todd D. The survival of glucose-6-phosphate dehydrogenase--deficient erythrocytes in patients with typhoid fever on chloramphenicol therapy. J Lab Clin Med 1971; 77:177.
  57. Constantopoulos A, Economopoulos P, Kandylas J. Fulminant diarrhoea and acute haemolysis due to G.-6-P.D. deficiency in salmonellosis. Lancet 1973; 1:1522.
  58. Hersko C, Vardy PA. Haemolysis in typhoid fever in children with G-6-PD deficiency. Br Med J 1967; 1:214.
  59. Mengel CE, Metz E, Yancey WS. Anemia during actue infections. Role of glucose-6-phosphate dehydrogenase deficiency in Negroes. Arch Intern Med 1967; 119:287.
  60. Salen G, Goldstein F, Haurani F, Wirts CW. Acute hemolytic anemia complicating viral hepatitis in patients with glucose-6-phosphate dehydrogenase deficiency. Ann Intern Med 1966; 65:1210.
  61. Baehner RL, Nathan DG, Castle WB. Oxidant injury of caucasian glucose-6-phosphate dehydrogenase-deficient red blood cells by phagocytosing leukocytes during infection. J Clin Invest 1971; 50:2466.
  62. Gellady AM, Greenwood RD. G-6-PD hemolytic anemia complicating diabetic ketoacidosis. J Pediatr 1972; 80:1037.
  63. Shalev O, Wollner A, Menczel J. Diabetic ketoacidosis does not precipitate haemolysis in patients with the Mediterranean variant of glucose-6-phosphate dehydrogenase deficiency. Br Med J (Clin Res Ed) 1984; 288:179.
  64. Glader BE. Role of elevated glucose concentrations in the hemolysis of glucose-6-phosphate dehydrogenase deficient erythroycytes (38474). Proc Soc Exp Biol Med 1975; 148:50.
  65. Johnson L, Bhutani VK, Karp K, et al. Clinical report from the pilot USA Kernicterus Registry (1992 to 2004). J Perinatol 2009; 29 Suppl 1:S25.
  66. Kaplan M, Hammerman C. Glucose-6-phosphate dehydrogenase deficiency: a hidden risk for kernicterus. Semin Perinatol 2004; 28:356.
  67. Kaplan M, Algur N, Hammerman C. Onset of jaundice in glucose-6-phosphate dehydrogenase-deficient neonates. Pediatrics 2001; 108:956.
  68. Kaplan M, Hammerman C, Feldman R, Brisk R. Predischarge bilirubin screening in glucose-6-phosphate dehydrogenase-deficient neonates. Pediatrics 2000; 105:533.
  69. Liu H, Liu W, Tang X, Wang T. Association between G6PD deficiency and hyperbilirubinemia in neonates: a meta-analysis. Pediatr Hematol Oncol 2015; 32:92.
  70. Sansone G, Perroni L, Yoshida A. Glucose-6-phosphate dehydrogenase variants from Italian subjects associated with severe neonatal jaundice. Br J Haematol 1975; 31:159.
  71. Kaplan M, Abramov A. Neonatal hyperbilirubinemia associated with glucose-6-phosphate dehydrogenase deficiency in Sephardic-Jewish neonates: incidence, severity, and the effect of phototherapy. Pediatrics 1992; 90:401.
  72. Huang CS, Hung KL, Huang MJ, et al. Neonatal jaundice and molecular mutations in glucose-6-phosphate dehydrogenase deficient newborn infants. Am J Hematol 1996; 51:19.
  73. Brown WR, Boon WH. Hyperbilirubinemia and kernicterus in glucose-6-phosphate dehydrogenase-deficient infants in Singapore. Pediatrics 1968; 41:1055.
  74. Perkins RP. The significance of glucose-6-phosphate dehydrogenase deficiency in pregnancy. Am J Obstet Gynecol 1976; 125:215.
  75. Slusher TM, Vreman HJ, McLaren DW, et al. Glucose-6-phosphate dehydrogenase deficiency and carboxyhemoglobin concentrations associated with bilirubin-related morbidity and death in Nigerian infants. J Pediatr 1995; 126:102.
  76. Gibbs WN, Gray R, Lowry M. Glucose-6-phosphate dehydrogenase deficiency and neonatal jaundice in Jamaica. Br J Haematol 1979; 43:263.
  77. Oyebola DD. Care of the neonate and management of neonatal jaundice as practised by Yoruba traditional healers of Nigeria. J Trop Pediatr 1983; 29:18.
  78. Drew JH, Kitchen WH. Jaundice in infants of Greek parentage: the unknown factor may be environmental. J Pediatr 1976; 89:248.
  79. Kaplan M, Vreman HJ, Hammerman C, et al. Contribution of haemolysis to jaundice in Sephardic Jewish glucose-6-phosphate dehydrogenase deficient neonates. Br J Haematol 1996; 93:822.
  80. Seidman DS, Shiloh M, Stevenson DK, et al. Role of hemolysis in neonatal jaundice associated with glucose-6 phosphate dehydrogenase deficiency. J Pediatr 1995; 127:804.
  81. Valaes T. Severe neonatal jaundice associated with glucose-6-phosphate dehydrogenase deficiency: pathogenesis and global epidemiology. Acta Paediatr Suppl 1994; 394:58.
  82. Valaes T. Pathophysiology of spontaneous neonatal bilirubinemia associated with glucose-6-phosphate dehydrogenase deficiency. J Pediatr 1996; 128:863.
  83. Drew JH, Smith MB, Kitchen WH. Glucose-6-phosphate dehydrogenase deficiency in immigrant greek infants. J Pediatr 1977; 90:659.
  84. Brown AK, Cevik N. Hemolysis and jaundice in the newborn following maternal treatment with sulfamethoxypyridazine (kynex). Pediatrics 1965; 36:742.
  85. Mentzer WC, Collier E. Hydrops fetalis associated with erythrocyte G-6-PD deficiency and maternal ingestion of fava beans and ascorbic acid. J Pediatr 1975; 86:565.
  86. Kaplan M, Rubaltelli FF, Hammerman C, et al. Conjugated bilirubin in neonates with glucose-6-phosphate dehydrogenase deficiency. J Pediatr 1996; 128:695.
  87. Kaplan M, Beutler E, Vreman HJ, et al. Neonatal hyperbilirubinemia in glucose-6-phosphate dehydrogenase-deficient heterozygotes. Pediatrics 1999; 104:68.
  88. MacDonald MG. Hidden risks: early discharge and bilirubin toxicity due to glucose 6-phosphate dehydrogenase deficiency. Pediatrics 1995; 96:734.
  89. Beutler E, Mathai CK, Smith JE. Biochemical variants of glucose-6-phosphate dehydrogenase giving rise to congenital nonspherocytic hemolytic disease. Blood 1968; 31:131.
  90. Beutler E, Grooms AM, Morgan SK, Trinidad F. Chronic severe hemolytic anemia due to G-6-PD Charleston: a new deficient variant. J Pediatr 1972; 80:1005.
  91. Feldman R, Gromisch DS, Luhby AL, Beutler E. Congenital nonspherocytic hemolytic anemia due to glucose-6-phosphate dehydrogenase East Harlem: a new deficient variant. J Pediatr 1977; 90:89.
  92. Grossman A, Ramanathan K, Justice P, et al. Congenital nonspherocytic hemolytic anemia associated with erythrocyte glucose-6-phosphate dehydrogenase deficiency in a Negro family. Pediatrics 1966; 37:624.
  93. Rattazzi MC, Corash LM, van Zanen GE, et al. G6PD deficiency and chronic hemolysis: four new mutants--relationships between clinical syndrome and enzyme kinetics. Blood 1971; 38:205.
  94. Vives Corrons JL, Feliu E, Pujades MA, et al. Severe-glucose-6-phosphate dehydrogenase (G6PD) deficiency associated with chronic hemolytic anemia, granulocyte dysfunction, and increased susceptibility to infections: description of a new molecular variant (G6PD Barcelona). Blood 1982; 59:428.
  95. https://www.fda.gov/drugs/science-research-drugs/table-pharmacogenomic-biomarkers-drug-labeling (Accessed on May 13, 2019).
  96. https://www.mayocliniclabs.com/test-catalog/Clinical+and+Interpretive/8368 (Accessed on March 16, 2021).
  97. https://ltd.aruplab.com/Tests/Pub/0080135 (Accessed on March 16, 2021).
  98. http://www.axismarcom.com/Files/Stanford/esoteric/RBC-Special-Studies.html (Accessed on March 16, 2021).
  99. Beutler E. Red cell metabolism. In: A Manual of Biochemical Methods, 3rd ed, Grune and Stratton, New York 1984.
  100. Motulsky AG, Yoshida A. Methods for the study of red cell, glucose-6-phosphate dehydrogenase. In: Biochemical Methods in Red Cell Genetics, Yunis JJ (Ed), Academic Press, New York 1969.
  101. Doherty AN, Kring EA, Posey YF, Maisels MJ. Glucose-6-phosphate dehydrogenase activity levels in white newborn infants. J Pediatr 2014; 164:1416.
  102. Grobusch MP, Rodríguez-Morales AJ, Schlagenhauf P. The Primaquine Problem-and the Solution? Point-of-care Diagnostics for Glucose 6-Phosphate Dehydrogenase Deficiency. Clin Infect Dis 2019; 69:1443.
  103. Baird JK. Point-of-care G6PD diagnostics for Plasmodium vivax malaria is a clinical and public health urgency. BMC Med 2015; 13:296.
  104. Bhutani VK, Kaplan M, Glader B, et al. Point-of-Care Quantitative Measure of Glucose-6-Phosphate Dehydrogenase Enzyme Deficiency. Pediatrics 2015; 136:e1268.
  105. Kaplan M, Hammerman C. The need for neonatal glucose-6-phosphate dehydrogenase screening: a global perspective. J Perinatol 2009; 29 Suppl 1:S46.
  106. Balinsky D, Gomperts E, Cayanis E, et al. Glucose-6-phosphate dehydrogenase Johannesburg: a new variant with reduced activity in a patient with congenital non-spherocytic haemolytic anaemia. Br J Haematol 1973; 25:385.
  107. Hafez M, Amar ES, Zedan M, et al. Improved erythrocyte survival with combined vitamin E and selenium therapy in children with glucose-6-phosphate dehydrogenase deficiency and mild chronic hemolysis. J Pediatr 1986; 108:558.
  108. Brewer GJ, Zarafonetis CJ. The haemolytic effect of various regimens of primaquine with chloroquine in American Negroes with G6PD deficiency and the lack of an effect of various antimalarial suppressive agents on erythrocyte metabolism. Bull World Health Organ 1967; 36:303.
  109. Lee SW, Lai NM, Chaiyakunapruk N, Chong DW. Adverse effects of herbal or dietary supplements in G6PD deficiency: a systematic review. Br J Clin Pharmacol 2017; 83:172.
  110. SZEINBERG A, SHEBA C, ADAM A. Enzymatic abnormality in erythrocytes of a population sensitive to Vicia faba or haemolytic anemia induced by drugs. Nature 1958; 181:1256.
  111. SANSONE G, SEGNI G. [New aspects of the biochemical alterations in the erythrocytes of patients with favism; almost complete absence of glucose-6-phosphate dehydrogenase]. Boll Soc Ital Biol Sper 1958; 34:327.
  112. http://www.g6pd.org (Accessed on May 31, 2013).
  113. Galiano S, Gaetani GF, Barabino A, et al. Favism in the African type of glucose-6-phosphate dehydrogenase deficiency (A-). BMJ 1990; 300:236.
  114. Kattamis CA, Kyriazakou M, Chaidas S. Favism: clinical and biochemical data. J Med Genet 1969; 6:34.
  115. Stamatoyannopoulos G, Fraser GR, Motulsky AC, et al. On the familial predisposition to favism. Am J Hum Genet 1966; 18:253.
  116. Cutillo S, Costa S, Vintuleddu MC, Meloni T. Salicylamide-Glucuronide formation in children with favism and in their parents. Acta Haematol 1976; 55:296.
  117. Belsey MA. The epidemiology of favism. Bull World Health Organ 1973; 48:1.
  118. Chevion M, Navok T, Glaser G, Mager J. The chemistry of favism-inducing compounds. The properties of isouramil and divicine and their reaction with glutathione. Eur J Biochem 1982; 127:405.
  119. Razin A, Hershko A, Glaser G, Mager J. The oxidant effect of isouramil on red cell glutathione and its synergistic enhancement by ascorbic acid or 3,4-dihydroxyphenylalanine. Possible relation to the pathogenesis of favism. Isr J Med Sci 1968; 4:852.
  120. Gaetani GF, Rolfo M, Arena S, et al. Active involvement of catalase during hemolytic crises of favism. Blood 1996; 88:1084.
Topic 7111 Version 46.0

References