Classifications in Brief: Laboratory Risk Indicator for Necrotizing Fasciitis Score

History

Necrotizing fasciitis is a life-threatening infection that requires a multidisciplinary approach to management, often involving orthopaedic surgeons for surgical débridement. It was first defined by Meleney [15] in 1924, when he discovered beta-hemolytic Streptococcus in a series of 20 patients with necrotizing soft-tissue infections, appropriately termed hemolytic Streptococcus gangrene. In 1952, Wilson [23] renamed the disease necrotizing fasciitis based on consistently present necrotic fascial planes, and this is the currently used term.

Since the original description, several classification systems have been used in patients with necrotizing fasciitis. These classifications have been based on anatomic location [13], causative organisms [9], depth of invasion [22], and surgical management of necrotizing fasciitis [2, 18]. However, these systems often address the clinical presentation and management of necrotizing fasciitis at a late stage, and thus, in our opinion, tend not to be useful in detecting early necrotizing fasciitis.

In 2004, Wong et al. [25] proposed a new scoring system based on commonly acquired laboratory values that enabled the detection and differentiation of early necrotizing fasciitis from other soft-tissue infections. This is widely known as the Laboratory Risk Indicator for Necrotizing Fasciitis (LRINEC) score. Despite its original purpose, a retrospective cohort study has also demonstrated that LRINEC scores greater than or equal to 6 are associated with an increased risk of death or amputation [20].

However, although the LRINEC score was meant to improve the ability of clinicians to diagnose this life- and limb-threatening condition early, studies have suggested that it is insufficiently sensitive for this purpose [1, 11]. Until, or unless, it is refined and demonstrated to be both reproducible and sensitive, clinicians probably should not use this score in practice without augmenting it with other diagnostic approaches, such as plain radiographs, CT, or MRI [1, 22].

Purpose

Necrotizing fasciitis is widely recognized as a rare yet deadly soft-tissue infection. The risk of death once a patient develops this condition is as high as 25% to 50%, earning its designation as an orthopaedic surgical emergency [10, 12, 19]. Given the high risk of mortality with necrotizing fasciitis, early detection is critical because early surgical débridement has been found to be the single most important modifiable factor to decrease mortality [19, 25]. Evaluation of patients who may have necrotizing fasciitis and subsequent surgical débridements often are performed by orthopaedic surgeons, given that the condition often occurs in patients with orthopaedic trauma and after extremity surgery [11, 12].

The LRINEC score is potentially attractive because it requires only routinely ordered laboratory values. The original score was developed using a retrospective approach that compared frequently acquired laboratory values in patients with confirmed necrotizing fasciitis to patients with abscesses or cellulitis [25]. The goal of the LRINEC score was to develop a noninvasive and objective scoring system that could help detect and distinguish early necrotizing fasciitis from other soft-tissue infections, given the similarity of the initial clinical presentation.

However, as subsequent studies have shown [1, 11], it appears not to be as sensitive as a classification must be for a potentially deadly condition.

Description

In the original retrospective study by Wong et al. [25], six criteria were found to allow for risk stratification of probable necrotizing fasciitis among patients with abscesses and cellulitis. These factors included C-reactive protein, white blood cell count, hemoglobin, sodium, creatinine, and blood glucose levels. Each of these factors were independently assigned a value from 0 to 4 based on the closest integer of their regression coefficients as categorical variables that independently predicted necrotizing fasciitis. Total scores were calculated by summing the numerical values of each of the six factors to give a total score between 0 and 13. The total LRINEC scores were then stratified into three groups based on the probability of necrotizing fasciitis and applied to a validation cohort. LRINEC scores of 6 or greater were seen in 89.9% to 92.9% of patients with necrotizing fasciitis compared with only 3.1% to 8.4% of control patients, and thus, scores of at least 6 were considered to indicate a moderate risk raising the suspicion of necrotizing fasciitis. Scores of at least 8 were considered to indicate a high risk and were strongly predictive of necrotizing fasciitis (Table 1) [25].

Table 1.

Laboratory Risk Indicator for Necrotizing Fasciitis (LRINEC) score [25]

Validation

The LRINEC score has been extensively evaluated regarding its ability to reliably detect early necrotizing fasciitis and differentiate it from other soft-tissue infections; however, studies have varied widely in their conclusions as to whether the LRINEC score is sufficiently sensitive to use in practice [11, 16, 17]. It is important to note that the majority of evidence concerning the applicability of the LRINEC score is of lower quality given that they are retrospective observational studies and case reports [1].

The original description of the LRINEC score cited a positive predictive value of 92% and a negative predictive value of 96% [25]. One study at a tertiary care center in Australia found similar values, with a positive predictive value of 95.5% and negative predictive value of 88.1% [16]. However, this particular study is limited because it did not consider the time to presentation and variable access to healthcare in Australia, which may result in late presentations that are more robustly reflected in the laboratory values of the LRINEC score [16].

Further studies have attempted to reproduce these findings with little success. A wide variation in the positive predictive value (25% to 64%), negative predictive value (42% to 100%), sensitivity (43.2% to 100%), and specificity (60% to 96.9%) was found in subsequent systematic reviews [1, 6, 11]. The clinical utility of the LRINEC score to diagnose early necrotizing fasciitis is therefore in question, given the wide range in sensitivity and specificity that has been reported. In fact, during development of the original LRINEC score, approximately 10% of patients with necrotizing fasciitis had a LRINEC score less than 6 [25]. Neeki et al. [17] performed a 10-year retrospective study in the emergency department setting and similarly found a high false-positive and high-false negative rate of the LRINEC score. Around 1/10 (10.7%) of patients with cellulitis were considered to be at a moderate or high risk of having necrotizing fasciitis based on their LRINEC scores, whereas more than half (63.3%) of patients with necrotizing fasciitis were considered to have a low risk based on their LRINEC scores (Table 1) [17].

Factors that have been found to influence the predictive value of the LRINEC score include the causative organism, the presence of diabetes, and the status of the patient’s immune system [1, 6, 11, 17]. One systematic review found a correlation between the Necrotizing Fasciitis Classification’s sensitivity and that of the LRINEC score [9, 11]. The Necrotizing Fasciitis Classification groups patients based on the causative organism, with Type 1 involving polymicrobial infections, Type 2 being predominantly monomicrobial and involving Group A Streptococcus with or without Staphylococcus, and Type 3 involving gram-negative marine bacteria such as Vibrio vulnificus [9, 10, 21]. The LRINEC score had the highest sensitivity in Types 1 and 2 infections (54.8% to 97%) compared with Type 3 infections (11% to 17%) [11]. Additionally, the presence of diabetes, a common comorbidity of necrotizing fasciitis, has been shown to increase the sensitivity and accuracy of the LRINEC score compared with people without diabetes [5, 6, 11, 17]. On the other hand, immunocompromise, especially in relation to hematologic malignancies, has been associated with decreased sensitivity of the LRINEC score because of leukopenia and thrombocytopenia at baseline [8].

Limitations

When dealing with a life- and limb-threatening condition, an early diagnostic test must be sensitive. Many retrospective studies [5, 8, 14, 17] and systematic reviews [1, 7, 11] have suggested that the LRINEC scale is not sufficiently sensitive to be used for this important task.

Thus, the use of the LRINEC score in clinical practice alone may result in missed diagnoses of necrotizing fasciitis. In fact, a LRINEC score of less than 6 was seen in anywhere from 43.3% to 63.8% of patients who were ultimately given a diagnosis of necrotizing fasciitis [14, 17]. One case report even described a LRINEC score of 0 in a healthy middle-aged man without diabetes who later had surgically confirmed necrotizing fasciitis, further demonstrating the low sensitivity of the LRINEC score [24].

Until this important limitation is addressed with modifications to the LRINEC score that improve its sensitivity, or unless other reliable diagnostic modalities such as CT with contrast [7] are used alongside it, the LRINEC score should not be used in practice.

There are several additional limitations to the LRINEC score, including the lack of a prospective evaluation and other diagnostic modalities that may have superior sensitivity compared with the LRINEC score. To the best of our knowledge, no prospective trials have evaluated the LRINEC score’s ability to distinguish necrotizing fasciitis as it was proposed by Wong et al. [25]. A retrospective evaluation of the LRINEC score has also failed to validate the clinical utility of the scoring system [17], and only a few studies have been able to reproduce the high positive and negative predictive values reported in Wong’s original study [14, 16].

Compared with other diagnostic modalities of necrotizing fasciitis, the LRINEC score does not show superior ability to detect early necrotizing fasciitis. When the LRINEC score (at least 6) was compared with plain radiographic studies and CT, the sensitivities were 68.2%, 48.9%, and 88.5%, respectively [7]. The corresponding specificities were 84.8%, 94.0%, and 93.3%, respectively [7]. The classically described physical examination findings (fever, hemorrhagic bullae, and hypotension) also had poor sensitivity for diagnosing necrotizing fasciitis [7]. The sensitivities of each of these tests are low, and the clinical suspicion of necrotizing fasciitis remains an invaluable tool for early diagnosis.

Modifications have been proposed to improve these limitations and increase the clinical utility of the LRINEC score. One study sought to increase the ability of the LRINEC score to differentiate early necrotizing fasciitis from other soft-tissue infections by creating a modified LRINEC score that replaced serum sodium and glucose levels with erythrocyte count and fibrinogen levels and incorporated four different clinical parameters (pain, fever, tachycardia, and signs of acute renal injury) [4]. This modified LRINEC score was found to increase the score’s positive predictive value without decreasing specificity. However, to the best of our knowledge, it has not been validated by others yet.

Conclusions

The LRINEC score should not be used in clinical practice alone due to the risk of missing the limb- and life-threatening diagnosis of necrotizing fasciitis.

Initially proposed as a promising and widely available diagnostic tool, the LRINEC score has several shortcomings that have prevented it from becoming the clinically reliable and practical tool it was expected to be. The LRINEC score alone has insufficient sensitivity to diagnose early cases of necrotizing fasciitis or to be used to rule-out the diagnosis of necrotizing fasciitis.

Given the generally low sensitivity and specificity of the LRINEC score, modifications to the score, such as incorporating clinical features, have been proposed to improve its accuracy, but this has yet to be studied extensively [3, 4]. Additional imaging studies, including CT with contrast or MRI have shown to have remarkably higher sensitivity and are thus are essential in detecting and differentiating necrotizing fasciitis [7].

Further studies should focus on validating the modified LRINEC score as proposed by Borschitz et al. [4] and evaluating the use of CT or MRI in combination with the LRINEC score. A high degree of clinical suspicion by the orthopaedic surgeon combined with a consideration of the patient comorbidities, clinical symptoms, laboratory values, and imaging techniques remains paramount to the early diagnosis and management of necrotizing fasciitis.

The LRINEC score adds minimal value as a diagnostic tool for identifying early cases of necrotizing fasciitis, and orthopaedic surgeons must use it with caution because a missed diagnosis may be fatal.