Neither baseline tests of molecular hypercoagulability nor D-dimer levels predict deep venous thrombosis in critically ill medical-surgical patients.

OBJECTIVE: Predicting patients who are harboring asymptomatic deep venous thrombosis (DVT), or who are at particular risk of developing DVT, is a desirable clinical goal since prevention or early treatment of DVT might reduce the risk of fatal pulmonary embolism. Thus validation of simple laboratory tests that reliably predict venous thromboembolism (VTE) would be clinically very important. Tests that might be useful for these applications include markers of hypercoagulability (predicting patients at risk of DVT) and D-dimer (predicting which patients may have acute DVT). METHODS: In a prospective cohort study we measured a panel of hypercoagulability markers at the time of ICU admission, and six commercial D-dimer assays were performed serially during the ICU stay in medical-surgical ICU patients who were screened for DVT with biweekly lower limb compression ultrasonography. Ultrasonography was also performed at the time of any clinically suspected DVT events. We matched cases with DVT with controls without DVT for length of stay in the ICU to generate receiver operating characteristics (ROC) curves. RESULTS: One hundred ninety-seven patients were enrolled. Blood was collected on a total of 763 occasions (median number of occasions per patient: 3, range 1-21). None of the assays predicted DVT, as indicated by the areas under the ROC curves, that did not differ significantly from 50%. CONCLUSION: In critically ill patients, neither tests of hypercoagulability nor D-dimer levels predict patients at risk of DVT and thus they should not be used to guide diagnostic testing for DVT.

Benefit of continuous renal replacement therapy in subgroups of acutely ill patients: a retrospective analysis.

AIMS: Acute renal failure in the intensive care setting is common and impacts on patient's outcome. Continuous hemodialysis or hemofiltration offers theoretical benefit for patients with acute renal failure, but the clinical benefit has not been demonstrated in randomized trials. ICU patients with acute renal failure are a heterogeneous population and we hypothesize that patients with increased illness severity would benefit from continuous renal replacement therapy. METHODS: From a comprehensive ICU database, we identified patients with acute renal failure exposed to continuous or intermittent renal replacement therapy. We a priori identified a subgroup of patients with multiple organ dysfunction syndrome, then used survival analysis to assess the effect of dialysis modality in the overall acute renal failure population and in the subgroup with increased illness severity. RESULTS: We identified 66 patients treated with intermittent and 36 patients treated with continuous renal replacement therapy. Patients with severe illness were preferentially selected for treatment with continuous dialysis (p = 0.01). Overall, there was no significant difference in survival between patients treated with intermittent or continuous dialysis. The relative risk of in-hospital mortality was significantly decreased in patients with multiple organ dysfunction syndrome (relative risk = 0.42+/-0.22, p = 0.027) treated with continuous therapy as compared with intermittent therapy. Among the survivors, continuous dialysis did not appear to hasten the return of renal function. CONCLUSIONS: This retrospective study suggests that continuous dialysis may be beneficial in a subgroup of ICU patients with multiple organ dysfunction syndrome or severe sepsis. Further randomized trials of dialysis modality should, if possible, concentrate on this population.

Peritoneal membrane transport and hypoalbuminemia: cause or effect?

OBJECTIVE: Peritoneal membrane transport has been associated with serum albumin and clinical outcome. We examined the relationship between serum albumin and peritoneal membrane transport status before and after the initiation of peritoneal dialysis. SETTING: Patients were followed at a tertiary-care regional nephrology program at St. Joseph's Hospital, McMaster University, Hamilton, Ontario, Canada. METHODS: Incident peritoneal dialysis patients between 1 January 1995 and 31 May 1998 were eligible if there was a peritoneal equilibration test within 180 days of starting dialysis, and a serum albumin value measured within 90 days prior to, and 20 to 70 days after initiating dialysis. MAIN OUTCOME MEASURES: Serum albumin, before and after the initiation of dialysis, and the presence of proteinuric renal disease were compared with the peritoneal equilibration test results. RESULTS: Among 67 identified patients, there were 7 high, 27 high-average, 26 low-average, and 7 low transporters and the mean serum albumin values before dialysis were 35.1, 37.4, 37.8, and 40.4 g/L, respectively (p < 0.001). Serum albumin values prior to the initiation of dialysis correlated significantly with the 4-hour D/P creatinine ratio (r = -0.251, p = 0.040). After initiation of dialysis, the correlation was stronger (r= -0.447, p< 0.001). Serum albumin prior to initiation of dialysis was lower for those with proteinuric than nonproteinuric renal disease (36.4 g/L vs 38.8 g/L, p = 0.04). The trend to lower serum albumin in high transporters was seen in patients with both proteinuric and nonproteinuric renal disease. CONCLUSION: The association between higher peritoneal membrane transport and lower serum albumin is present before initiation of dialysis in both proteinuric and nonproteinuric renal disease. The poor outcomes associated with low serum albumin and higher peritoneal membrane transport might be explained by other underlying factors. The contribution of inflammation, malnutrition, and fluid overload requires further study.

Event adjudication and data monitoring in an intensive care unit observational study of thromboprophylaxis.

BACKGROUND: The objective of this report is to describe the roles, responsibilities and recommendations of a 3-member Event Adjudication Committee (EAC) and a 5-member data monitoring committee (DMC) for a prospective multicenter observational study of critically ill patients with renal insufficiency examining the bioaccumulation and bleeding risk associated with dalteparin thromboprophylaxis. METHODS: The EAC reviewed bleeding events to adjudicate whether they were major or minor and whether they were related to dalteparin (uncertain, unlikely, or likely). The DMC reviewed all bleeds deemed by the EAC as uncertain or likely due to dalteparin then recommended either to continue or suspend enrolment pending review by the steering committee, or requested more information. RESULTS: Consensus on bleeding severity was achieved for all cases. At the second planned interim analysis, the EAC recommended to the DMC and steering committee that the EAC should stop adjudicating whether bleeding was related to dalteparin since attribution was challenging in this population with numerous bleeding risk factors; moreover, no bleeding rates were available from prior studies or historical or concurrent controls. CONCLUSIONS: Adjudication of whether an outcome can be attributed to an intervention in an open-label, uncontrolled observational study gives a potentially misleading impression of research oversight without methodological face validity. In this study, the EAC recommended modification of the adjudication process, and the DMC recommended continuing enrolment to achieve the target sample size.

The clinical utility of Doppler ultrasound prior to arteriovenous fistula creation.

Arteriovenous fistula (AVF) is the preferred access for long-term hemodialysis, with superior long-term patency rates; however, early failure rates are significant. Recent evidence has brought into question the preferred site of AVF creation in many patient groups. A preoperative test that could reliably predict the outcome of a proposed AVF would be of great benefit. Doppler ultrasound has been the most extensively studied and widely used test to guide access creation. Accurate and validated measurements of internal vessel diameter, both arterial and venous, and blood flow in the upper extremity are obtainable by Doppler ultrasound. Studies evaluating the utility of Doppler ultrasound prior to AVF creation suggest that vessel size and blood flow are predictive of AVF outcome. An AVF created using a cephalic vein and/or radial artery smaller than 1.5-2.0 mm is likely to fail; such preoperative data may indicate that an upper arm AVF should be the primary access attempted. Further prospective studies are needed to evaluate the utility of Doppler ultrasound.