Portal de Periódicos da CAPES

The Diagnosis of Acute Mesenteric Ischemia: A Systematic Review and Meta-analysis

2013; Wiley; Volume: 20; Issue: 11 Linguagem: Inglês

10.1111/acem.12254

1553-2712

Michael T. Cudnik, Subrahmanyam Darbha, J B Jones, Julian Macedo, Sherrill W. Stockton, Brian Hiestand,

Gallbladder and Bile Duct Disorders

Acute mesenteric ischemia is an infrequent cause of abdominal pain in emergency department (ED) patients; however, mortality for this condition is high. Rapid diagnosis and surgery are key to survival, but presenting signs are often vague or variable, and there is no pathognomonic laboratory screening test. A systematic review and meta-analysis of the available literature was performed to determine diagnostic test characteristics of patient symptoms, objective signs, laboratory studies, and diagnostic modalities to help rule in or out the diagnosis of acute mesenteric ischemia in the ED. In concordance with published guidelines for systematic reviews, the medical literature was searched for relevant articles. The Quality Assessment Tool for Diagnostic Accuracy Studies-2 (QUADAS-2) for systematic reviews was used to evaluate the overall quality of the trials included. Summary estimates of diagnostic accuracy were computed by using a random-effects model to combine studies. Those studies without data to fully complete a two-by-two table were not included in the meta-analysis portion of the project. The literature search identified 1,149 potentially relevant studies, of which 23 were included in the final analysis. The quality of the diagnostic studies was highly variable. A total of 1,970 patients were included in the combined population of all included studies. The prevalence of acute mesenteric ischemia ranged from 8% to 60%. There was a pooled sensitivity for l-lactate of 86% (95% confidence interval [CI] = 73% to 94%) and a pooled specificity of 44% (95% CI = 32% to 55%). There was a pooled sensitivity for D-dimer of 96% (95% CI = 89% to 99%) and a pooled specificity of 40% (95% CI = 33% to 47%). For computed tomography (CT), we found a pooled sensitivity of 94% (95% CI = 90% to 97%) and specificity of 95% (95% CI = 93% to 97%). The positive likelihood ratio (+LR) for a positive CT was 17.5 (95% CI = 5.99 to 51.29), and the negative likelihood ratio (–LR) was 0.09 (95% CI = 0.05 to 0.17). The pooled operative mortality rate for mesenteric ischemia was 47% (95% CI = 40% to 54%). Given these findings, the test threshold of 2.1% (below this pretest probability, do not test further) and a treatment threshold of 74% (above this pretest probability, proceed to surgical management) were calculated. The quality of the overall literature base for mesenteric ischemia is varied. Signs, symptoms, and laboratory testing are insufficiently diagnostic for the condition. Only CT angiography had adequate accuracy to establish the diagnosis of acute mesenteric ischemia in lieu of laparotomy. La isquemia mesentérica aguda es una causa infrecuente de dolor abdominal en los servicios de urgencias (SU); sin embargo la mortalidad para esta enfermedad es alta. El diagnóstico rápido y la cirugía son clave para la supervivencia, pero los signos de presentación son a menudo vagos o variables, y no hay una prueba de despistaje de laboratorio patognomónica. Se realizó una revisión sistemática y un metanálisis de la literatura disponible para determinar las características diagnósticas de los síntomas del paciente, los signos objetivos, los estudios de laboratorio y las modalidades diagnósticas para ayudar a descartar el diagnóstico de isquemia mesentérica aguda en el SU. En consonancia con las guías clínicas publicadas para revisiones sistemáticas, la bibliografía médica se buscó para los artículos relevantes. Se uso Quality Assessment Tool for Diagnostic Accuracy Studies-2 (QUADAS-2) para las revisiones sistemáticas para evaluar la calidad global de los estudios incluidos. Las estimaciones resumidas de certeza diagnóstica se calcularon mediante un modelo de efectos aleatorios para combinar estudios. Aquellos estudios sin datos para completar totalmente una tabla de dos por dos no se incluyeron en la parte del metanálisis del estudio. La búsqueda bibliográfica identificó 1.149 estudios potencialmente relevantes, de los cuales 23 se incluyeron en el análisis final. La calidad de los estudios diagnósticos fue muy variable. Un total de 1.970 pacientes se incluyeron en la población combinada de todos los estudios incluidos. La prevalencia de isquemia aguda mesentérica osciló de un 8% a un 60%. Hubo una sensibilidad acumulada para el lactato de un 86% (IC 95% = 73% a 94%), y una especificidad acumulada de un 44% (IC 95% = 32% a 55%). Hubo una sensibilidad acumulada para el D-dímero de un 96% (IC 95% = 89% a 99%), y una especificidad acumulada de un 40% (IC 95% = 33% a 47%). Para la tomografía computarizada (TC), se encontró una sensibilidad acumulada de un 94% (IC 95% = 90% a 97%) y especificidad de un 95% (IC 95% = 93% a 97%). La razón de probabilidad positiva para una TC positiva fue de 17,5 (IC 95% = 5,99 a 51,29) y la razón de probabilidad negativa de 0,09 (IC 95% = 0,05 a 0,17). El porcentaje de mortalidad acumulada para la isquemia mesentérica fue de un 47% (IC 95% = 40% a 54%). Dados estos hallazgos, se calculó el umbral diagnóstico de un 2,1% (por debajo de esta probabilidad pre-test, no hacer más pruebas diagnósticas), y un umbral de tratamiento de un 74% (por encima de esta probabilidad pre-test, proceder a un tratamiento quirúrgico). La calidad de la bibliografía global para la isquemia mesentérica es variada. Los signos, síntomas y pruebas de laboratorio son insuficientes para diagnosticar esta enfermedad. Sólo la angio-TC tuvo una adecuada certeza para establecer el diagnóstico de isquemia mesentérica aguda en lugar de la laparotomía. CME Editor: Hal Thomas, MD Authors: Michael T. Cudnik, MD, Subrahmanyam Darbha, MSc, Janice Jones, MD, Julian Macedo, MD, Sherrill W. Stockton, MD, PhD, and Brian C. Hiestand, MD, MPH Article Title: The Diagnosis of Acute Mesenteric Ischemia: A Systematic Review and Meta-analysis If you wish to receive free CME credit for this activity, please refer to the website: http://www.wileyhealthlearning.com/aem. Accreditation and Designation Statement: Blackwell Futura Media Services designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM. Physicians should only claim credit commensurate with the extent of their participation in the activity. Blackwell Futura Media Services is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. Educational Objectives After completing this exercise the participant will better understand the diagnosis of acute mesenteric ischemia. Activity Disclosures No commercial support has been accepted related to the development or publication of this activity. Faculty Disclosures: CME editor – Hal Thomas, MD: No relevant financial relationships to disclose. Authors Michael T. Cudnik, MD, Subrahmanyam Darbha, MSc, Janice Jones, MD, Julian Macedo, MD, Sherrill W. Stockton, MD, PhD, and Brian C. Hiestand, MD, MPH. This manuscript underwent peer review in line with the standards of editorial integrity and publication ethics maintained by Academic Emergency Medicine. The peer reviewers have no relevant financial relationships. The peer review process for Academic Emergency Medicine is double-blinded. As such, the identities of the reviewers are not disclosed in line with the standard accepted practices of medical journal peer review. Conflicts of interest have been identified and resolved in accordance with Blackwell Futura Media Services's Policy on Activity Disclosure and Conflict of Interest. No relevant financial relationships exist for any individual in control of the content and therefore there were no conflicts to resolve. Instructions on Receiving Free CME Credit For information on applicability and acceptance of CME credit for this activity, please consult your professional licensing board. This activity is designed to be completed within an hour; physicians should claim only those credits that reflect the time actually spent in the activity. To successfully earn credit, participants must complete the activity during the valid credit period, which is up to two years from initial publication. Follow these steps to earn credit: This activity will be available for CME credit for twelve months following its publication date. At that time, it will be reviewed and potentially updated and extended for an additional twelve months. Acute mesenteric ischemia is a rare disease, with an annual incidence of 0.09% to 0.2% per patient year, although the disease is thought to be underreported.1-3 Accordingly, acute mesenteric ischemia is an infrequent cause of abdominal pain in emergency department (ED) patients; however, mortality for this condition is high.4-7 Rapid diagnosis and surgical intervention are paramount to limiting mortality, but presenting signs are often vague or variable, and there is no pathognomonic laboratory screening test.8-11 Even after the clinical suspicion for mesenteric ischemia is acknowledged, definitive diagnosis has traditionally required invasive and time-consuming subtraction angiography or specialized computed tomography (CT) techniques in conjunction with expert radiologic interpretation.12-14 Accordingly, despite a growing clinical awareness and a rapid advancement of laboratory assays and radiologic techniques generally, the timely diagnosis of acute mesenteric ischemia remains challenging.9, 15, 16 Mortality rates vary in different published series, with a range of roughly 50% to 100%, likely secondary to varying disease etiologies (location of infarct, venous vs. arterial blockage, etc.).17-19 Mesenteric ischemia is typically caused by arterial thrombi to the celiac axis, the superior mesenteric artery, or the inferior mesenteric artery, leading to decreased intestinal blood flow and direct ischemic and secondary reperfusion cellular damage.20 This is occlusive mesenteric arterial ischemia, which accounts for approximately half of all reported cases.21 Roughly one-third of cases are nonocclusive mesenteric arterial ischemia and result from vasoconstriction, low output states, or both.21 Most of the remaining cases of mesenteric ischemia are the result of mesenteric venous thrombosis.21 Although mortality after intervention may vary depending on the precise etiology, ranging from a reported 32% for mesenteric venous thrombosis to 77% for arterial thrombosis,19 the emergency physician must be tasked with considering the global condition of “acute mesenteric ischemia” as a cause for the patient's presentation, rather than specific subtypes. Acute mesenteric ischemia is a different clinical entity than chronic mesenteric ischemia or ischemic colitis. Chronic mesenteric ischemia occurs due to gradual stenosis, usually secondary to atherosclerotic disease, of the arterial supply to the viscera.22, 23 Pain, characterized as intestinal angina, is generally postprandial. Although debilitating due to pain and weight loss, chronic mesenteric ischemia is not acutely fatal and is not considered further within this review. Likewise, colonic ischemia refers to a heterogeneous collection of presentations that manifest as transmural colonic ischemia, usually due to a nonocclusive but low blood flow state insufficient to meet the metabolic demands of the colonic tissues. This may be due to vasospasm or systemic hypotension, frequently in the setting of an atherosclerotic arterial supply.24 This is a distinct clinical entity from acute mesenteric ischemia as well and is not the topic of this review. Regardless of the specific etiology, early diagnosis is central to successful management of mesenteric ischemia. The intestine does enjoy significant collateral circulation throughout its course, and it can tolerate a 75% reduction in blood flow for up to 12 hours.25 This is clearly demonstrated in the mesenteric ischemia literature where the relationship between early diagnosis and intervention and decreased mortality is well established.9, 26, 27 Despite the acknowledged importance of early diagnosis of mesenteric ischemia, the goal remains difficult to achieve, which may contribute to the fact that mortality rates for acute mesenteric ischemia have remained consistently poor over time.4, 6 A major challenge in the diagnosis of mesenteric ischemia is the wide spectrum of patient presentations. Symptoms vary from the classic “pain out of proportion to physical exam,” to vague or insidious abdominal symptoms, to absent abdominal pain.1, 28, 29 Another diagnostic challenge is the lack of adequate laboratory markers. Numerous candidate plasma markers have been studied, among them serum lactic acid dehydrogenase, D-dimer, ischemia-modified albumin, and urinary and plasma fatty acid–binding proteins (FABPs).16, 30-34 Another diagnostic difficulty involves imaging techniques. At many centers, direct angiography has been supplanted by multidetector row CT as the initial imaging technique of choice for the diagnosis of suspected mesenteric ischemia. In short, there remains no laboratory test, imaging technique, or risk stratification tool with adequate sensitivity and specificity to effectively rule out or rule in patients in whom mesenteric ischemia is a concern. The primary objective of this meta-analysis was to assess the diagnostic test characteristics for acute mesenteric ischemia from elements of patient symptoms, objective signs, laboratory studies, or imaging studies in ED patients. A secondary objective was to define mesenteric ischemia imaging test and treatment thresholds using the Pauker-Kassirer method based on best estimates of sensitivity, specificity, diagnostic risks, and treatment risks and benefits that were derived from this systematic literature review.35 The design and structure of this systematic review followed the recommendations from the Meta-analysis of Observational Studies in Epidemiology.36 Three investigators (SWS, JJ, JM) searched the medical literature from 1966 through December 2011 using PUBMED and EMBASE using the search term acute mesenteric ischemia. The results from this search were combined with the MeSH terms diagnosis, labs, emergency department, computed tomography, and angiography. To identify the risks of intravenous (IV) dye for CT and IV dye from angiography for determination of the test threshold analysis, a PUBMED search was done using the terms angiography and risk, angiography and complications, CT and risk, and CT and complications. To identify the risks and benefits of operative or nonoperative management of acute mesenteric ischemia, a PUBMED clinical query under “therapy” was done. All search results were limited to studies of humans and English language. Two authors (MTC, BCH) reviewed the titles of the abstracts to identify potential articles for inclusion and reviewed the full manuscripts. These two authors independently reviewed each of the articles for potential inclusion. Consensus was achieved via discussion if there was a difference of opinion between authors regarding article inclusion. The references for selected articles were also reviewed to identify other potential articles for inclusion. Finally, one author (JJ) searched online for abstracts and articles in Academic Emergency Medicine and Annals of Emergency Medicine from 1990 to 2011. Studies were included if they included adult patients (≥18 years old) who presented to the hospital or ED with suspicion of acute mesenteric ischemia and if they reported sufficient data on diagnostic tests and criterion standard results to reconstruct two-by-two tables in whole or in part. Case reports, narrative reviews, and studies focusing on therapy alone were excluded. Two authors (MTC, BCH) used the Quality Assessment Tool for Diagnostic Accuracy Studies-2 (QUADAS-2) for systematic reviews to evaluate the overall quality of the trials included in the meta-analysis.37 Any discrepancies of the quality assessments were resolved by discussion. Statistical agreement between the two reviewers was assessed via kappa analysis using Online Kappa Calculator.38 A priori, the authors considered potential areas of concern with the assessment of the articles. Both ED and hospital-based populations were included in the study. If a trial did not explicitly state that investigators were blinded to the index test and/or the reference standard, then these portions of the QUADAS-2 were marked as “high bias.” For those articles with relevant cohort information regarding symptom or physical examination prevalence (i.e., only disease-positive patients were studied), formal bias assessment with QUADAS-2 was not performed. In the absence of disease-negative patients, true diagnostic accuracy cannot be assessed, and we did not intend to include these studies in the diagnostic meta-analysis or test–treatment threshold calculations. They are included for descriptive purposes only. Two authors (MTC, BCH) independently abstracted the data from the included studies. Data abstracted included setting, patient population, study inclusion criteria, reference standard employed, disease prevalence, and properties of the respective diagnostic tests. A priori, we defined disease as acute mesenteric ischemia, as proven by operative findings or autopsy findings. We defined no disease as the absence of acute mesenteric ischemia as evidenced by clinical resolution of symptoms without intervention or negative operative findings. We then computed summary estimates by combining study patients with and without diagnoses of mesenteric ischemia using Meta-DiSc39 (Hospital Universitario Ramon y Cajal, Madrid, Spain) using the Der-Simonian random-effects model.40 Meta-DiSc was also used to generate summary receiver operator characteristic (SROC) curves. SROC curves provide graphical summaries of diagnostic data performance in meta-analyses, providing a summary overall diagnostic odds ratio as well as incorporating interstudy heterogeneity in the graphical output.41 Those studies without data to fully complete a two-by-two table were not included in the calculation of diagnostic test characteristics (sensitivity, specificity, likelihood ratios), although they could contribute to prevalence analysis via simple pooling of results. Statistical heterogeneity was assessed for pooled estimates via the Cochrane's χ2 and I2 statistic with 25, 50, and 75% representing low, moderate, and high heterogeneity, respectively.42 The PUBMED search identified 1,037 citations, while the EMBASE search identified 382 (Figure 1). No additional studies were obtained after reviewing abstracts and articles from the two emergency medicine journals. After initial screening, 87 unique manuscripts were selected for potential inclusion. After full manuscript review, a total of 23 studies were included in the final meta-analysis on diagnostic testing.1, 21, 28-30, 32-34, 43-57 These 23 studies consisted of 17 prospective studies1, 21, 30, 32, 34, 43-45, 47-49, 51, 53-57 and six retrospective studies.28, 29, 33, 46, 50, 52 In terms of history and physical examination findings, 19 studies were able to contribute prevalence data and estimates of test sensitivity.1, 4-7, 9, 10, 28, 29, 50, 58-66 A summary of all studies contributing data can be reviewed in Data Supplement S1 (available as supporting information in the online version of this paper). The studies included a variety of patients in both the hospital and the ED settings with a wide range of inclusion criteria and diagnostic modalities including elements of the history and physical examination, laboratory tests, and radiographic imaging. A large number of the studies only included those patients with diagnoses of mesenteric ischemia, therefore lacking a control or comparator group. Only one study34 acknowledged the STARD criteria67 for diagnostic studies; however, several studies included in this analysis were published prior to the release of the STARD criteria in 2003. The reliability for the authors’ QUADAS-2 assessments of quality ranged from kappas of 0.52 to 0.88. The quality of the diagnostic studies was highly variable (Table 1). Only five studies explicitly stated that they included ED patients.29, 34, 47, 54, 57 Many studies did not describe the inclusion criteria in detail. Additionally, the vast majority of studies did not report the precise intervals between the index test and reference standard. Most of the studies reporting laboratory test results used dichotomous cutoff values for continuous variables of diagnostic tests (lactate, D-dimer). A total of 1,970 patients were included in the combined population of all included studies. The prevalence of mesenteric ischemia ranged from 8%56 to 60%44, 49 in those studies that included patients with and without mesenteric ischemia. The summary median age of patients was 67 years (interquartile range = 63 to 70 years). Risk factors and corresponding sensitivity ranges from the history, clinical presentation, and physical examination are listed in Table 2. Given the lack of a comparator group without mesenteric ischemia in the studies included, no data on the specificity or the likelihood ratios could be calculated, limiting the diagnostic utility of these historical features. A history of atrial fibrillation was frequently present in the setting of mesenteric ischemia (sensitivity range = 7.7% to 79.3%). Abdominal pain was also frequently present in patients with mesenteric ischemia, ranging from 60% to 100% prevalence in the case series. Physical examination findings suggestive of mesenteric ischemia were diffuse abdominal tenderness (sensitivity range = 54% to 90%), the presence of peritoneal signs (sensitivity range = 13% to 65%), and abdominal distention (sensitivity range = 18% to 54%). However, none of the studies reported findings on all patients with suspected mesenteric ischemia to calculate specificity or likelihood ratios, again limiting the diagnostic utility of these findings. Four studies evaluated the diagnostic accuracy of elevated lactate levels in patients with suspected mesenteric ischemia. The studies used varying cutoffs, which are reported with each study. Murray et al.53 (d-lactate ≥ 2.0 μg/mL) reported a sensitivity of 89% and a specificity of 86%, while Block et al.32 (D-lactate ≥ 0.20 mmol/L) reported a sensitivity of 90% with a specificity of 23%. This led to a pooled sensitivity for d-lactate of 90% (95% confidence interval [CI] = 67% to 99%) and a pooled specificity of 40% (95% CI = 29% to 51%). The heterogeneity was low for sensitivity (I2 = 0%, Cochran's Q χ21 = 0.01, p = 0.937) but quite high for specificity (I2 = 96.5%, Cochran's Q χ21 = 28.3, p < 0.001). It should be noted that the d-lactate isomer is a product of bacterial metabolism, as opposed to l-lactate, which is a product of human anaerobic metabolism. Both isomers will contribute to metabolic acidosis; however, a specific assay is required for the detection of d-lactate. Gearhart et al.49 (l-lactate ≥ 2.2 mmol/L) reported sensitivity and specificity of 78 and 53% respectively, while data from Lange and Jackel52 demonstrated a sensitivity of 100% (but with a lower limit of the 95% CI of 83%) and a specificity of 42%, using a cutoff of 2.4 mmol/L. These studies led to a pooled sensitivity for l-lactate of 86% (95% CI = 73% to 94%) and a pooled specificity of 44% (95% CI = 32% to 55%). With regard to l-lactate, high heterogeneity was noted between trials for sensitivity (I2 = 87.5%, Cochran's Q χ21 = 7.99, p = 0.005) and low heterogeneity for specificity (I2 = 0%, Cochran's Q χ21 = 0.66, p = 0.416). Data Supplement S2 provides forest plots for d-lactate, and Data Supplement S3 provides forest plots for l-lactate. SROC curves were not calculated for d- or l-lactate due to too few studies. Five studies evaluated the diagnostic accuracy of elevated D-dimers in mesenteric ischemia patients.30, 32, 43, 44, 47 The studies used varying cutoffs for defining an elevated D-dimer. All reported sensitivities of ≥95%, with three reporting sensitivities of 100%. The overall specificity of an elevated D-dimer ranged from 18% to 79%. All of the studies reported significantly higher D-dimer levels in patients with mesenteric ischemia compared to those without mesenteric ischemia. These studies led to a pooled sensitivity of 96% (95% CI = 89% to 99%) and a pooled specificity of 40% (95% CI = 33% to 47%). As seen in Table 3, the positive likelihood ratio (+LR) for an elevated d-dimer was 1.76 (95% CI = 1.20 to 2.57), and the negative likelihood ratio (–LR) was 0.12 (95% CI = 0.05 to 0.30). Little heterogeneity was noted between trials for sensitivity (I2 = 0%, Cochran's Q χ24 = 0.66, p = 0.96), but significant heterogeneity was observed between trials for specificity (I2 = 86%, Cochran's Q χ24 = 29.02, p < 0.001). Data Supplement S4 provides forest plots for D-dimer, and Data Supplement S5 provides the SROC curve. Alpha-glutathione S-transferase (GST) and intestinal FABPs, two tests not readily available in most EDs, were also evaluated. The sensitivity of GST ranged from 20% to 100% in the three studies evaluating it.32, 48, 49 Additionally, the sensitivity of FABP ranged from 64% to 100% depending on the study and which isomer of FABP was used.33, 34 Data Supplement S6 provides forest plots for GST, while Data Supplement S7 provides the SROC curve. Data Supplement S8 presents forest plots for FABP. An SROC could not be calculated for FABP due to too few studies. The most common test used for diagnosis of mesenteric ischemia was CT angiography. No studies meeting our inclusion criteria were found that assessed magnetic resonance imaging. Overall, there were eight studies that investigated the sensitivity and specificity of CT for the diagnosis of mesenteric ischemia.21, 45, 46, 51, 54, 56, 57, 66 The studies used different types of scanners in their studies (4-row to 64-row scanners), with one study46 using three different types of scanners (16, 40, and 64 rows) on their study population. Sensitivity ranged from 83% to 100% with a pooled sensitivity of 94% (95% CI = 90% to 97%), while specificity ranged from 67% to 100% with a pooled specificity of 95% (95% CI = 93% to 97%). As seen in Table 3, the +LR for a positive CT was 17.5 (95% CI = 5.99 to 51.29) and the –LR for a negative CT scan was 0.09 (95% CI = 0.05 to 0.17). Moderate heterogeneity was noted between trials for sensitivity (I2 = 59.5%, Cochran's Q χ27 = 17.29, p = 0.016) and high levels of heterogeneity were noted in terms of specificity (I2 = 87.6%, Cochran's Q χ27 = 56.27, p < 0.001). Data Supplement S9 presents forest plots for CT imaging, and Data Supplement S10 provides the SROC curve. The most common management option of acute mesenteric ischemia is surgical exploration and intervention. There is a substantial degree of heterogeneity in the literature with regard to reported postoperative mortality after surgical intervention for suspected acute mesenteric ischemia. The operative mortality of mesenteric ischemia ranged from 26% to 72% with a pooled mortality rate of 47% (95% CI = 40% to 54%). Operative repair of mesenteric ischemia was associated with significant morbidity and adverse event rates in survivors, ranging from 39% to 64%. However, the mortality rate for missed mesenteric ischemia that is not operated on is considered to approach 100% Acute mesenteric ischemia remains a difficult diagnosis to establish on a clinical basis. There are no presenting characteristics, historical features, or findings on physical examination that definitively establish the diagnosis. Rather, the pretest probability of disease must be gradually informed by the accumulation of mild shifts in likelihood provided by the presence or absence of various elements in the history and physical. As of yet, there are no laboratory tests readily available in the ED that possess enough diagnostic accuracy to establish mesenteric ischemia definitively. An ideal screening test would have a very low –LR. In general, a –LR less than 0.1 is considered an adequate “proof of absence of disease,” as the posttest odds of disease would be decreased by a factor of 10.68 A negative D-dimer has a strong –LR, although the 95% CI was high enough that we cannot recommend that it is a satisfactorily accurate screening test to preclude further testing as a standalone result. Serum lactate, the diagnostic test most frequently associated with mesenteric ischemia, had a lower pooled sensitivity than D-dimer and a higher –LR. Studies of these two tests did not consistently establish the interval between specimen acquisition and the final diagnosis of mesenteric ischemia. As well, neither test was substantially specific. This is in evidence by the wide CIs about the SROC (Data Supplement S5). It may also be that there are different degrees of predictive ability from different degrees of abnormality of the laboratory assay. In other words, a patient with a very elevated lactate may carry a higher probability of acute mesenteric ischemia than a patient who is just over the upper limit of normal. The heterogeneity of the source data precluded our ability to investigate this further; future prospective work may find incremental value in calculating interval likelihood ratios to evaluate this possibility. We found substantially less variability with regard to CT imaging. However, there were outliers in terms of both sensitivity and specificity. Wiesner et al.56 performed a retrospective review of abdominal CT interpretations performed for a variety of indications and with a variety of protocols on a four-slice multidetector CT, which may have contributed to their substantially lower sensitivity than the rest of the field. Likewise, Blachar et al.46 performed a retrospective review of images that were reinterpreted for study purposes in a case–control format, allowing CT images that were acquired up to 3 days prior to the diagnosis of mesenteric ischemia. CT scans were performed on a variety of multidetector scanners (ranging from 10-to 64-slice) throughout the study