INTRODUCTION — Diabetic muscle infarction, which is also referred to as spontaneous diabetic myonecrosis, is the term used for spontaneous ischemic necrosis of skeletal muscle, unrelated to atheroembolism or occlusion of major arteries. It causes acute or subacute pain, swelling, and tenderness, typically in the thigh or calf.
Diabetic muscle infarction is one of many micro- and macrovascular complications of diabetes. Others include diabetic retinopathy, nephropathy, neuropathy, and atherosclerotic vascular disease affecting other circulatory beds.
The clinical manifestations, diagnosis, differential diagnosis, treatment, and prognosis of diabetic muscle infarction are discussed here. Other musculoskeletal manifestations of diabetes mellitus and disorders with an increased prevalence in patients with diabetes are presented separately, as is an approach to minimizing the risk of complications arising from diabetic micro- and macrovascular disease. (See "Overview of the musculoskeletal complications of diabetes mellitus" and "Overview of general medical care in nonpregnant adults with diabetes mellitus".)
PATHOGENESIS — The pathogenesis of diabetic muscle infarction is uncertain but appears related to vasculopathic changes associated with longstanding and poorly controlled diabetes; factors responsible for the acute onset of symptoms and findings are unknown. The primary pathologic findings in muscle biopsies from affected patients are muscle necrosis and edema; occlusion of arterioles and capillaries by fibrin may also be seen [1,2].
EPIDEMIOLOGY — Spontaneous infarction of muscle is a rare condition; it occurs in both type 1 and type 2 diabetes, and the majority of patients have other multiple microvascular complications, including retinopathy, nephropathy, and/or neuropathy [1,3-5]. The prevalence of this condition is difficult to determine as most information has been obtained from analyses of published case reports.
A 2015 systematic review of all cases published in the English language since the initial description of the condition in 1965 identified reports of 170 episodes of diabetic muscle infarction, including 126 initial episodes and 44 episodes of recurrence, with a mean age at presentation of 45 years (range 20 to 67 years) [5].
The number of patients with type 1 and type 2 diabetes identified in this review were similar, but the mean age at presentation with muscle infarction was lower for type 1 diabetics than for type 2 diabetics (36 years [20 to 65 years] versus 52 years [34 to 67 years]). The mean duration of diabetes at the time of presentation with muscle infarction was longer for those with type 1 diabetes (19 years [5 to 33 years] versus 11 years [1 to 25 years]).
Other complications of diabetes were present in the majority of patients, with the most common microvascular complication, nephropathy, present in 75 percent of the patients. Of the 126 cases reviewed, 47 percent had concurrent retinopathy, nephropathy, and neuropathy, and 66 percent had at least two diabetic complications.
These findings were similar to those in earlier reviews, including one with fewer patients (n = 116) that included non-English-language reports as well [4]. In this 2003 review, the oldest reported patient was 81 years of age, and rates of vascular complications of diabetes were similar to the 2015 report, but patients with type 1 diabetes were more common than those with type 2 diabetes (59 versus 24 percent, with 17 percent unknown).
A systematic review of 41 cases (published from 1980 to 2017) of patients with diabetic muscle infarction and end-stage renal failure due to diabetic nephropathy showed similar findings and incidence of diabetic complications to earlier reviews, in which some of these case reports will have been included [6]. In this review, patients had a mean age of 44.2 years (19 to 67 years), 41.5 percent had type 1 diabetes, 81 percent were on dialysis, and 12 percent had kidney transplants. The recurrence rate for diabetic muscle infarction in this group was 43.9 percent.
CLINICAL MANIFESTATIONS
Symptoms and physical findings — Affected patients present with the acute onset of painful swelling that then evolves over days or weeks. The commonest area affected is the front of the thigh (55 percent), with the next most commonly affected areas being the back of the thigh or calf (15 percent) [5]. There is no history of trauma. The swelling may be mildly to extremely tender, and a minority of patients (approximately 10 percent [4,5]) are described as having a mild fever [1,3], although the majority of case reports do not record body temperature [4]. Erythema, induration, and warmth are described in some reports [3,7].
The mean duration of symptoms before presenting for care is about four weeks but ranged from within one day to 40 weeks in one review, and time to resolution ranged from 2 to 17 weeks, with a mean of four weeks [4]. In some cases (approximately one-third), a palpable painful mass may develop after partial resolution [4].
In the largest review, bilateral involvement occurs at presentation in about 8 percent of patients [5]. Recurrence is common in published cases but often in a different location [5,7]. In the 2015 review, 170 episodes of diabetic muscle infarction were reported in 126 patients, and 61 percent of these recurrences were in a different location or muscle group from the original site [5]. In one case report, for example, a 61-year-old type 2 diabetic with left thigh muscle infarction developed muscle infarction of the contralateral thigh five weeks after the initial episode, followed by muscle infarction in the right calf three months later [7]. The frequency of recurrence is uncertain, and the possibility of publication bias has to be considered as recurrent cases may be more likely to be reported. Uncommonly, the upper limb may be involved. In the smaller review of patients with end-stage renal failure and diabetic muscle infarction [6], on presentation 19.5 percent of patients had bilateral involvement and 39 percent involvement of multiple muscle groups. In this group, lower limb involvement predominated, with the thigh affected four times more frequently than the leg below the knee. Upper limb involvement occurred in 5 out of 41 patients.
Other complications, including compartment syndromes and secondary infections, are infrequently reported in patients with this rare condition. As an example, an apparently spontaneous compartment syndrome requiring surgical intervention has also been described in patients in whom a diabetic muscle infarction is suspected as the underlying cause [8] (see "Acute compartment syndrome of the extremities"). This case report and literature review identified nine patients with type 1 diabetes aged 17 to 47 years (median age 29 years) with spontaneous compartment syndrome, of whom about one-half had diabetic complications such as retinopathy. In addition, neck muscle (levator scapulae) involvement, later complicated by staphylococcal sepsis in the necrotic muscle and secondary osteomyelitis, has been described in an immunosuppressed diabetic transplant patient [9].
Laboratory — Laboratory findings are nonspecific and are normal in many patients. However, patients often exhibit elevated levels of creatine kinase (CK) and elevated acute phase responses, as well as leukocytosis. Based upon data from two large literature reviews, which were partially overlapping, the following results were observed [5,7]:
●CK levels were normal or increased, probably depending upon the stage of the condition when sampling was undertaken. CK levels were greater than 150 units (U)/L, the upper limit of normal, in 32 to 47 percent of patients, with a mean value of 710 U/L and values ranging from 10 to 11,000 U/L.
●Leukocytosis has been found in 35 to 43 percent of patients in whom the leukocyte count was reported, with mean values in one report of 11.5 x 109/L, but levels seen as high as 32 x 109/L.
●Elevation of the erythrocyte sedimentation rate (ESR) to more than 50 mm/hour was seen in 74 percent of patients in whom it was reported in one study but was measured in less than one-half of the patients [7]. Similarly, in the other report, the ESR was elevated in 83 percent of patients in whom it was reported, with a mean value for all patients reported of 87 mm/hour (range 1 to 153 mm/hour).
●Only 10 percent of patients who had their C-reactive protein (CRP) reported had a normal CRP, with mean values for all patients reported of 156 mg/L (range 0.03 to 524 mg/L).
●Measurements of glycated hemoglobin (HbA1C) were also only performed in a minority (40 percent) of patients, and in this study the mean value was 9.3 percent [5].
Imaging features — Magnetic resonance imaging (MRI) and ultrasonography have each been used to assess patients with possible diabetic muscle infarction. Their uses and relative utility for diagnosis and differential diagnosis are described separately. (See 'Diagnosis' below and 'Differential diagnosis' below.)
MRI — MRI may show high intensity in the involved muscle on T2-weighted sequences as well as subcutaneous edema and subfascial fluid [3,7,10]. The loss of the normal fatty intramuscular septa, a relatively common finding, is optimally observed with T1-weighted images, where the affected muscles appear hypointense or isointense.
The addition of gadolinium can distinguish nonenhancing infarcted muscle from surrounding inflammation or edema. On post-gadolinium scans, there is diffuse heterogenous enhancement with low-signal, nonenhancing foci that may represent areas of necrosis. Rim enhancement can be seen around these areas of necrosis within the areas of ischemic muscle [11,12].
Ultrasonography — There is a paucity of data regarding ultrasonography findings. In one study of patients with diabetic muscle infarction, the presence of internal linear echogenic structures, the absence of a predominant anechoic area, and no evidence of internal motion or swirling fluid with transducer pressure could discriminate a diabetic muscle infarction from an abscess [13].
Arteriography — Arteriography, which is not generally performed for diagnostic purposes in patients with suspected diabetic muscle infarction, may reveal atherosclerotic luminal narrowing [14]. There are very few studies of this modality in this context.
Muscle biopsy — Muscle necrosis and edema are the primary pathologic findings, but occlusion of arterioles and capillaries by fibrin may also be seen [1,2]. (See 'Pathogenesis' above.)
DIAGNOSIS — Awareness of the syndrome and the presence of the characteristic clinical features, usually including acute atraumatic swelling, warmth, and pain in a typically affected area such as the thigh, will frequently suggest the diagnosis. Laboratory and imaging studies are aimed at excluding other disorders; laboratory changes suggesting an inflammatory process may be present, and an inflammatory response to the muscle infarction may make differentiation from infection difficult.
Magnetic resonance imaging (MRI) with intravenous contrast enhancement appears to be the most useful diagnostic imaging technique and is the diagnostic imaging tool of choice. (See 'Differential diagnosis' below.)
Muscle biopsy is not required for the diagnosis of muscle infarction and is only necessary when the diagnosis remains in doubt despite careful clinical assessment, laboratory tests, and appropriate imaging. Muscle biopsy is usually performed only in patients in whom infection cannot be excluded by other investigative techniques. Since MRI clearly identifies the location of muscle involvement, it may be useful to guide prebiopsy assessment [10].
For patients with longstanding diabetes, known microvascular complications, no fever, and no leukocytosis, the findings on MRI of increased T2 signal, muscle enlargement, and subcutaneous and subfascial edema may avoid muscle biopsy [7,11].
Diagnostic evaluation
Testing in all patients — We suggest the following diagnostic testing in all patients with suspected diabetic muscle infarction:
●Complete blood count (CBC) and differential.
●Prothrombin time and partial tissue thromboplastin time.
●Creatine kinase (CK), C-reactive protein (CRP), and erythrocyte sedimentation rate (ESR).
●Blood cultures for aerobic and anaerobic bacteria to exclude infection.
●Plain film radiography of the affected area to assess for gas in affected muscle to exclude clostridial myonecrosis.
●Venous Doppler ultrasound with compression to exclude deep vein thrombosis.
●MRI of affected and contralateral limbs. The MRI can usually be performed without contrast, but the need for gadolinium contrast or other enhancing agents can be assessed on a case-by-case basis after discussion with the radiologist, taking into account the clinical presentation and comorbidities, particularly the patient's renal function (given the high prevalence of nephropathy in these patients), and the noncontrast MRI findings.
A retrospective study of MRI imaging from 16 patients with diabetic muscle infarction found that the use of gadolinium-based contrast agents may not be necessary in diagnosing patients with a typical presentation of diabetic muscle infarction [15]. The addition of gadolinium can distinguish nonenhancing infarcted muscle from surrounding inflammation or edema [11]; however, administration of gadolinium-containing MRI contrast agents should be avoided in patients with moderately or severely impaired renal function (eg, estimated glomerular filtration rate <15 to 30 mL/minute). If gadolinium-based imaging must be performed in a patient with moderately to severely impaired renal function, there may be a role for dialysis to reduce the risk of development of nephrogenic systemic fibrosis (NSF). Causes of and approaches to prevention of NSF are discussed separately. (See "Patient evaluation before gadolinium contrast administration for magnetic resonance imaging", section on 'Approach to preventing nephrogenic systemic fibrosis' and "Nephrogenic systemic fibrosis/nephrogenic fibrosing dermopathy in advanced kidney disease", section on 'Prevention'.)
Testing in selected patients — In patients suspected of muscle infarction in whom infection cannot be excluded, additional studies, including biopsy, may be required. Biopsy is rarely needed. For patients with fever and/or leukocytosis, in whom muscle infection cannot be excluded on the basis of the tests above, we recommend computed tomography (CT)-guided core needle biopsy and examination of the tissue with Gram stain, culture, and routine histopathology. Needle biopsy may provide diagnostic tissue and is less likely to exacerbate the disease or prolong recuperation than is excisional muscle biopsy, which may be associated with hematoma, delayed healing, and infection [16].
In patients in whom a rapidly progressive infection of muscle or fascia cannot be excluded or in those with signs of sepsis, we suggest surgical exploration and excisional biopsy. (See 'Differential diagnosis' below.)
Differential diagnosis — Among other causes of acute muscle pain and tenderness, bacterial infections such as pyomyositis, spontaneous gangrenous myositis, nontraumatic clostridial myonecrosis, and necrotizing fasciitis must be excluded expeditiously. Ultrasonography used to exclude venous thrombosis may reveal intramuscular hematomas or muscle tears. Diabetic patients with end-stage kidney disease may develop soft tissue infarction due to calciphylaxis.
●Pyomyositis – There is an increased propensity to pyomyositis in patients with diabetes mellitus, and large muscles of the thigh are a commonly involved site. Early pyomyositis may be difficult to distinguish from ischemic muscle necrosis; however, fever, leukocytosis, and a well-defined intramuscular fluid collection is commonly noted at a more advanced stage of bacterial muscle infection. (See "Pyomyositis".)
●Spontaneous gangrenous myositis – Spontaneous gangrenous myositis due to streptococcal infection, unlike pyomyositis due to Staphylococcus aureus, is characterized by gangrenous necrosis, rather than abscess formation; more systemic toxicity; and a much worse prognosis (80 to 100 percent mortality). (See "Necrotizing soft tissue infections".)
●Clostridial myonecrosis – The presence of gas in muscle is characteristic of nontraumatic clostridial myonecrosis. (See "Clostridial myonecrosis", section on 'Spontaneous gas gangrene'.)
●Necrotizing fasciitis – Necrotizing fasciitis is a deep-seated infection of the subcutaneous tissue that results in progressive destruction of fascia and fat. Necrotizing fasciitis should be considered in diabetic patients with cellulitis who also have systemic signs of infection such as tachycardia, leukocytosis, marked hyperglycemia, or acidosis. (See "Necrotizing soft tissue infections", section on 'Necrotizing fasciitis'.)
●Venous thrombosis – Pain, swelling, and skin discoloration in the involved extremity are common symptoms of venous thrombosis, and these may be accompanied by the presence of a palpable cord (reflecting a thrombosed vein), ipsilateral edema, warmth, and/or superficial venous dilation. Noninvasive studies, particularly compression ultrasonography, may distinguish venous thrombosis from muscle infarction. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity".)
●Intramuscular hematoma – Intramuscular bleeding due to low-energy blunt trauma or in patients with a heritable or acquired coagulopathy causes pain, swelling, and muscle tenderness. A history of trauma, bleeding disorder, or use of an anticoagulant is suggestive. Ischemic pain may result from increased intramuscular pressure that can be determined by measuring intracompartmental pressure [17]. Ultrasonography, CT, or MRI of the hematoma is usually diagnostic [18,19]. (See "Quadriceps muscle and tendon injuries" and "Quadriceps muscle and tendon injuries", section on 'Quadriceps contusion'.)
●Calciphylaxis – Calciphylaxis (calcific uremic arteriolopathy) is typically characterized by areas of ischemic necrosis that develop in the dermis, subcutaneous fat, and less often involve muscle. Typically seen in patients with end-stage kidney disease, these ischemic changes lead to livedo reticularis and/or violaceous, painful, plaque-like subcutaneous nodules on the trunk, buttocks, or proximal extremity. (See "Calciphylaxis (calcific uremic arteriolopathy)".)
●Neoplasm – Enlarging benign and malignant neoplasms are unlikely to be confused with muscle infarction, but tumor infarction or hemorrhage may cause acute pain and increasing size [7]. (See "Clinical presentation, histopathology, diagnostic evaluation, and staging of soft tissue sarcoma", section on 'Introduction'.)
TREATMENT — Treatment of diabetic muscle infarction involves symptomatic management with rest, optimal glycemic control, analgesia, and low-dose aspirin, although the optimal treatment approach is uncertain. There are very limited data suggesting that nonsteroidal antiinflammatory drug (NSAID) treatment may improve outcome, but convincing evidence is lacking; moreover, in this population with a high incidence of nephropathy, NSAIDs will often be contraindicated.
Our approach (algorithm 1A-B) is based upon published case reports and case series [1,5,7,20]; there have been no randomized trials to compare approaches or specific agents.
●We suggest low-dose aspirin (eg, 75 to 162 mg daily) for patients with diabetic muscle infarction who are not already receiving antiplatelet therapy and who have no contraindications to aspirin therapy, such as active peptic ulceration, aspirin-exacerbated respiratory disease (aspirin-induced asthma), and aspirin allergy. As recurrent episodes are frequent in the published literature, it is reasonable that aspirin should be continued indefinitely, although evidence for prevention is lacking.
In patients unable to take aspirin, another antiplatelet agent (eg, clopidogrel, 75 mg daily) may be employed. However, there are no studies to support the benefits of this approach specifically for diabetic muscle infarction, although it is used as an alternative antiplatelet agent in patients with diabetes mellitus and cardiovascular disease who are allergic to or otherwise unable to take aspirin but in whom aspirin would otherwise be indicated for the prevention of adverse cardiovascular events. (See "Overview of general medical care in nonpregnant adults with diabetes mellitus", section on 'Aspirin'.)
●Routine gentle daily activity, while often painful, is not harmful, but physiotherapy should be avoided during the acute phase of the illness, which typically lasts at least several weeks (see 'Symptoms and physical findings' above and 'Prognosis' below), as some studies suggest it may cause worsening of the condition and prolong the time to symptom resolution [20]. Physical activity can be increased after resolution of the acute symptoms, and physical therapy may be useful after discharge from the hospital but has not been evaluated prospectively [1,5,20].
●In the occasional diabetic patient with a muscle infarction but without contraindications, an NSAID (eg, naproxen 375 mg twice daily or ibuprofen 400 to 600 mg three times daily) can be used for short-term symptom relief in the lowest dose needed and for the least duration required, but risks (eg, risk factors for renal impairment or bleeding, established cardiovascular disease or high risk of such disease) must be carefully weighed against the benefits in this group of patients likely to be at higher risk of complications from the use of NSAIDs.
Although the use of NSAIDs has been proposed by some experts to speed recovery [5,7], the diabetic with muscle infarction is likely to have comorbidities and/or diabetic complications that are relative contraindications to treatment with these agents, such as renal disease, hypertension, heart failure, ischemic heart disease, and use of anticoagulants. Risk of peptic ulcer disease and gastrointestinal hemorrhage should also be assessed. (See "Nonselective NSAIDs: Overview of adverse effects".)
Concomitant use of aspirin and NSAIDs has the potential to cause adverse drug interactions. Among these are interference with the beneficial antiplatelet effect of aspirin and an increased risk of gastroduodenal damage. An approach to use of these combinations and strategies for mitigating the elevated gastrointestinal risk are described separately. (See "NSAIDs (including aspirin): Primary prevention of gastroduodenal toxicity" and "NSAIDs: Adverse cardiovascular effects", section on 'Aspirin and other antithrombotic agents'.)
●In patients who are at high risk for NSAID-related adverse effects and need to avoid these agents, we use non-NSAID analgesics (eg, acetaminophen). Those with severe muscle pain may require opioids (eg, codeine, tramadol, or other agents), but these should be used in the lowest dose necessary for the minimal duration of treatment that is required. (See "Pharmacologic management of chronic non-cancer pain in adults", section on 'Pharmacologic therapy for nociceptive pain'.)
PROGNOSIS — Diabetic muscle infarction resolves spontaneously over a few weeks to months in most patients. An analysis of previously published patient outcomes by the type of treatment received noted the following mean times to recovery [7]:
●Rest and analgesics – 8 weeks
●Antiplatelet agents and/or antiinflammatory drugs – 5.5 weeks
●Surgical excision – 13 weeks
These data are limited from being based upon a literature review of a number of case reports and case series without a common definition for recovery and likely confounding by indication.
Despite the initial symptomatic resolution and recovery, a second episode, often involving the contralateral side, has frequently been reported [1,3,5,7,20], with a risk of recurrence of approximately 40 percent [7]. The long-term outlook is poor, reflecting the underlying arteriopathy, with death from a major vascular event reported to occur within a few years in the majority of patients in a 1993 review [1]. A 2004 estimate of mortality within two years of an episode was 10 percent, suggesting some improvement in the prognosis of reported cases [7].
SUMMARY AND RECOMMENDATIONS
●Diabetic muscle infarction is a rare disorder that affects patients who have relatively longstanding diabetes, many of whom have other micro- or macrovascular complications. (See 'Epidemiology' above.)
●Clinical manifestations include acute or subacute onset of muscle pain, swelling, and associated tenderness. The muscles of the thigh and calf are most commonly affected. Frequent but nonspecific laboratory findings include elevation of plasma creatine kinase (CK) enzyme activity, leukocytosis, and elevated erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP), each occurring in a substantial proportion of patients. (See 'Symptoms and physical findings' above and 'Laboratory' above.)
●A clinically based diagnosis of muscle infarction may be appropriately made for patients with compatible magnetic resonance imaging (MRI) findings (ie, increased T2 signal in affected muscle, fascia, and subcutaneous tissues) and without clinical or laboratory features favoring infection (eg, fever, leukocytosis). Plain film radiography and venous compression ultrasonography are indicated to exclude the presence of gas in the soft tissues and assess for deep venous thrombosis. In patients in whom the diagnosis is uncertain, definitive diagnosis of diabetic muscle infarction requires biopsy of the affected area of muscle that demonstrates ischemic necrosis and excludes infection. (See 'Diagnosis' above.)
●We perform a computed tomography (CT)-guided core needle biopsy in patients with suspected muscle infarction who also have clinical or imaging features suggesting infection of muscle or fascia and in those in whom a malignant neoplasm cannot be excluded. (See 'Muscle biopsy' above and 'Diagnosis' above.)
●We perform surgical exploration and excisional biopsy in patients in whom a rapidly progressive infection of muscle or fascia cannot be excluded or in those with signs of sepsis. (See 'Muscle biopsy' above and 'Diagnosis' above.)
●In all patients with diabetic muscle infarction, we suggest low-dose aspirin as an antiplatelet agent to treat diabetic muscle infarction (algorithm 1A) (Grade 2C). Typically, we use a dose of 75 to 162 mg daily (see 'Treatment' above). Clopidogrel (75 mg daily) is an alternative antiplatelet agent for patients who are unable to take aspirin. (See 'Treatment' above and "Overview of general medical care in nonpregnant adults with diabetes mellitus", section on 'Aspirin'.)
●In the occasional diabetic patient with a muscle infarction but without risk factors for renal impairment or bleeding and without established cardiovascular disease or high risk of such disease, a nonsteroidal antiinflammatory drug (NSAID (algorithm 1B); eg, naproxen 375 mg twice daily or ibuprofen 400 to 600 mg three times daily) can be used for short-term symptom relief, but risks must be carefully weighed against the benefits in this group of patients likely to be at higher risk of complications from the use of NSAIDs. (See 'Treatment' above.)
●In patients unable to take NSAIDs who require pain relief, we use non-NSAID analgesics tailored to the needs and comorbidities of the individual patients (algorithm 1B). (See 'Treatment' above.)
