Primary fascial closure after damage control laparotomy: Sepsis vs haemorrhage

Objective

To compare the outcomes of patients undergoing damage control laparotomy (DCL) for intra-abdominal sepsis vs intra abdominal haemorrhage. We hypothesize that patients undergoing DCL for sepsis will have a higher rate of septic complications and a lower rate of primary fascial closure.

Settings and patients

Retrospective study of patients undergoing DCL from December 2006 to November 2009. Data are presented as medians and percentages where appropriate.

Results

111 patients were identified (55 men), 79 with sepsis and 32 with haemorrhage. There was no difference in age (63 vs 62 years), body mass index (BMI, 27 vs 28), diabetes mellitus (13% vs 9%), or duration of initial operation (125 vs 117 min). Patients with sepsis presented with a lower serum lactate (2.2 vs 4.7 mmol/L, p < 0.01), base deficit (4.0 vs 8.0, p ≤ 0.01) and ASA score (3.0 vs 4.0, p < 0.01). There was no statistical difference in overall morbidity (81% vs 66), mortality (19% vs 22%), intra-abdominal abscess (18% vs 16%), deep wound infection (9% vs 9%), enterocutaneous fistula (ECF) (8% vs 6%) and primary fascial closure (58% vs 59%). Multivariable analysis demonstrated that intra-abdominal abscess (OR 4.26, 95% CI 1.06–19.32), higher base deficit (OR 1.14, 95% CI 1.00–1.31) and more abdominal explorations (OR 1.54, 95% CI 1.23–2.07) were associated with lack of primary fascial closure, but BMI (OR 1.00, 95% CI 0.94–1.07), ECF (OR 2.02, 95% CI 0.23–19.98), wound infection (OR 0.93, 95% CI 0.15–5.27), amount of crystalloids infused within the first 24 h (OR 1.00, 95% CI 0.99–1.00) and intra-abdominal sepsis (OR 1.14, 95% CI 0.35–3.80) were not.

Conclusions

There was an equivalent rate of septic complications and primary fascial closure rates regardless of cause for DCL. Intra-abdominal abscess, worse base deficit and higher number of abdominal explorations were independently associated with the lack of primary fascial closure.     

Antibioprophylaxie pour césarienne avant incision ou après clampage du cordon ? Méta-analyse

Objective

The objective of this systematic review was to assess the effect of preoperative rather than after umbilical cord clamping antimicrobial prophylaxis for caesarean delivery on maternal and neonatal infectious postoperative morbidity.

Study design

Meta-analysis.

Methods

Three electronic databases (Pubmed, Cochrane Central Register of Randomized Controlled Trials and Embase) were searched without language restriction and retrieved 201 potentially relevant trials. Five randomized controlled trials (n = 1108) studying the timing of antimicrobial prophylaxis for caesarean section were included. The quality of included trials was assessed on the modified Oxford validity scale.

Results

Preoperative administration of antibiotics (n = 456) rather than after cord clamping (n = 563) provides a significant reduction in the incidence of endometritis (Odds Ratio (OR) 0.59 [95% Confidence Interval (CI) 0.35–0.98]) and of total maternal infectious morbidity (OR 0.51 [95% CI 0.32–0.82]). This benefit was not observed regarding the incidence of wound infection (Peto OR 0.58 [95% CI 0.29–1.16]), neonatal infection (Peto OR 1.06 [95% CI 0.57–1.96]), neonatal sepsis workup (OR 1.02 [95% CI 0.67–1.54]), neonatal documented sepsis (Peto OR 0.93 [95% CI 0.43–2.02]) or neonatal intensive care unit admission (OR 0.97 [95% CI 0.61–1.56]). No significant heterogeneity was observed between the included studies.

Conclusion

This meta-analysis provides strong evidence that the preoperative rather than after cord clamping administration of antimicrobial prophylaxis for caesarean delivery provides a reduction in the incidence of endometritis and maternal total infectious morbidity without affecting the incidence of wound infection and neonatal infectious morbidity.     

Systematic Review and Meta-Analysis of Robotic-Assisted versus Conventional Laparoscopic Pyeloplasty for Patients with Ureteropelvic Junction Obstruction: Effect on Operative Time,Length of Hospital Stay,Postoperative Complications,and Success Rate

Background

Although robotic-assisted procedures may theoretically be more advantageous than conventional laparoscopic ones, few studies have shown clear superiority of robotic-assisted laparoscopic pyeloplasty (RAP) over conventional laparoscopic pyeloplasty (CLP) for ureteropelvic junction obstruction (UPJO).

Objective

To undertake a systematic review and meta-analysis to evaluate the effect of RAP versus CLP for patients with UPJO, focusing on operative time, length of hospital stay, postoperative complications, and success rate.

Design, setting, and participants

We searched four electronic bibliographic databases, including the related articles PubMed feature, reference lists from articles, and program abstracts from scientific meetings. Consequently, 58 citations were identified. Two individuals independently screened the titles and abstracts of each citation to select the articles (90% agreement).

Intervention

Studies that compared RAP with CLP for treatment of UPJO were included. Case series on RAP or CLP were excluded because of large heterogeneity.

Measurements

We utilized weighted mean difference (WMD) to measure operative time and length of hospital stay and odds ratio (OR) and risk difference (RD) to measure complication and success rates. These ORs were pooled using a random effects model and were tested for heterogeneity.

Results

We identified eight publications that strictly met our eligibility criteria. Meta-analysis of extractable data showed that RAP was associated with a 10-min operative time reduction (WMD: −10.4 min; 95% CI: −24.6–3; p = 0.15) and significantly shorter hospital stay compared with CLP (WMD: −0.5 d; 95% CI: −0.6–−0.4; p < 0.01). There were no differences between the approaches with regard to rates of complication (OR: 0.7; 95% CI: 0.3–1.6; p = 0.40) and success (OR: 1.3; 95% CI: 0.5–3.5; p = 0.62).

Conclusions

RAP and CLP appear to be equivalent with regard to postoperative urinary leaks, hospital readmissions, success rates, and operative time.     

Risk factors for hypotension in urgently intubated burn patients

Background

When urgently intubating patient in the burn intensive care unit (BICU), various induction agents, including propofol, are utilized that may induce hemodynamic instability.

Methods

A retrospective review was performed of consecutive critically ill burn patients who underwent urgent endotracheal intubation in BICU. Basic burn-related demographic data, indication for intubation, and induction agents utilized were recorded. The primary outcomes of interest were clinically significant hypotension requiring immediate fluid resuscitation, initiation or escalation of vasopressors immediately after intubation. Secondary outcomes included ventilator days, stay length, and in-hospital mortality.

Results

Between January 2003 and August 2010, we identified 279 urgent intubations in 204 patients. Of these, the criteria for presumed sepsis were met in 60% (n = 168) of the intubations. After intubation, 117 patients (42%) experienced clinically significant hypotension. Propofol (51%) was the most commonly utilized induction agent followed by etomidate (23%), ketamine (15%), and midazolam (11%). On multiple logistic regression, %TBSA (OR 1.016, 95% CI 1.004–1.027, p < 0.001) and presumed sepsis (OR 1.852, 95% CI 1.100–3.117, p = 0.02) were the only significant predictors of hypotension. None of the induction agents, including propofol, were significantly associated with hypotension in patients with or without presumed sepsis.

Conclusions

In critically ill burn patients undergoing urgent endotracheal intubation, specific induction agents, including propofol, were not associated with clinically significant hypotension. Presumed sepsis and %TBSA were the most important risk factors.     

Developing a burn injury severity score (BISS): Adding age and total body surface area burned to the injury severity score (ISS) improves mortality concordance

Background

There is limited research validating the injury severity score (ISS) in burns. We examined the concordance of ISS with burn mortality. We hypothesized that combining age and total body surface area (TBSA) burned to the ISS gives a more accurate mortality risk estimate.

Methods

Data from the Royal Perth Hospital Trauma Registry and the Royal Perth Hospital Burns Minimum Data Set were linked. Area under the receiver operating characteristic curve (AUC) measured concordance of ISS with mortality. Using logistic regression models with death as the dependent variable we developed a burn-specific injury severity score (BISS).

Results

There were 1344 burns with 24 (1.8%) deaths, median TBSA 5% (IQR 2–10), and median age 36 years (IQR 23–50). The results show ISS is a good predictor of death for burns when ISS ≤ 15 (OR 1.29, p = 0.02), but not for ISS > 15 (ISS 16–24: OR 1.09, p = 0.81; ISS 25–49: OR 0.81, p = 0.19). Comparing the AUCs adjusted for age, gender and cause, ISS of 84% (95% CI 82–85%) and BISS of 95% (95% CI 92–98%), demonstrated superior performance of BISS as a mortality predictor for burns.

Conclusion

ISS is a poor predictor of death in severe burns. The BISS combines ISS with age and TBSA and performs significantly better than the ISS.     

The incidence and outcomes of acute kidney injury amongst patients admitted to a level I trauma unit

Purpose

This study aimed to identify the incidence and outcomes of patients with trauma related acute kidney injury (AKI), as defined by RIFLE criteria, at a single level I trauma centre and trauma ICU.

Methods

We performed a retrospective observational study of 666 patients admitted to a trauma ICU from a level I trauma unit from March 2008 to March 2011. We conducted multivariable logistic regression to identify independent predictors for AKI and mortality.

Results

The overall incidence of AKI was 15% (n = 102). Median injury severity score (ISS) was 25 (inter quartile range [IQR] 16–34) and mean age was 39 (SD 16.3) in the AKI group. Thirteen patients (13%) were referred with rhabdomyolysis associated renal Failure. Overall mortality in the AKI group was 57% (n = 58) but was significantly lower in the rhabdomyolysis Failure group (23% versus 64%; p = 0.012). AKI was independently associated with older age, base excess (BE) < −12 (odd ratio [OR] 22.9, 95% confidence interval [CI] 1.89–276.16), IV contrast administration (OR 2.7 95% CI 1.39–5.11) and blunt trauma (OR 2.2 95% CI 1.04–4.71). AKI was an independent predictor of mortality (OR 8.5, 95% CI 4.51–15.95). Thirty-nine (38%) patients required renal replacement therapy.

Conclusions

AKI in critically ill trauma patients is an independent risk factor for mortality and is independently associated with increasing age and low BE. Renal replacement therapy utilisation is high in this group and represents a significant health care cost burden.     

Updated Systematic Review and Meta-Analysis of the Comparative Data on Colposuspensions,Pubovaginal Slings,and Midurethral Tapes in the Surgical Treatment of Female Stress Urinary Incontinence

Context

Burch colposuspension, pubovaginal sling, and midurethral retropubic tape (RT) and transobturator tape (TOT) have been the most popular surgical treatments for female stress urinary incontinence (SUI). Several randomized controlled trials (RCTs) have been published comparing the different techniques, with conflicting results.

Objective

Our aim was to evaluate the efficacy, complication, and reoperation rates of midurethral tapes compared with other surgical treatments for female SUI.

Evidence acquisition

A systematic review of the literature was performed using the Medline, Embase, Scopus, Web of Science databases, and Cochrane Database of Systematic Reviews.

Evidence synthesis

Thirty-nine RCTs were identified. Patients receiving midurethral tapes had significantly higher overall (odds ratio [OR]: 0.61; confidence interval [CI]: 0.46–0.82; p = 0.00009) and objective (OR: 0.38; CI: 0.25–0.57; p < 0.0001) cure rates than those receiving Burch colposuspension, although they had a higher risk of bladder perforations (OR: 4.94; CI: 2.09–11.68; p = 0.00003). Patients undergoing midurethral tapes and pubovaginal slings had similar cure rates, although the latter were slightly more likely to experience storage lower urinary tract symptoms (LUTS) (OR: 0.31; CI: 0.10–0.94; p = 0.04) and had a higher reoperation rate (OR: 0.31; CI: 0.12–0.82; p = 0.02). Patients treated with RT had slightly higher objective cure rates (OR: 0.8;CI: 0.65–0.99; p = 0.04) than those treated with TOT; however, subjective cure rates were similar, and patients treated with TOT had a much lower risk of bladder and vaginal perforations (OR: 2.5; CI: 1.75–3.57; p < 0.00001), hematoma (OR: 2.62; CI: 1.35–5.08; p = 0.005), and storage LUTS (OR: 1.35; CI: 1.05–1.72; p = 0.02). Meta-analysis demonstrated similar outcomes for TVT-O (University of Liège, Liège, Wallonia, Belgium) and Monarc (AMS, Minnetonka, MN, USA).

Conclusions

Patients treated with RT experienced slightly higher continence rates than those treated with Burch colposuspension, but they faced a much higher risk of intraoperative complications. RT and pubovaginal slings were similarly effective, although patients with pubovaginal slings were more likely to experience storage LUTS. The use of RT was followed by objective cure rates slightly higher than TOT, but subjective cure rates were similar. TOT had a lower risk of bladder and vaginal perforations and storage LUTS than RT. The strength of these findings is limited by the heterogeneity of the outcome measures and the short length of follow-up.     

Laparoendoscopic Single-site Partial Nephrectomy: A Multi-institutional Outcome Analysis

Background

Laparoendoscopic single-site surgery (LESS) has been developed in an attempt to further reduce the surgical trauma associated with conventional laparoscopy. Partial nephrectomy (PN) represents a challenging indication for LESS.

Objective

To report a large multi-institutional series of LESS-PN and to analyze the predictors of outcomes after LESS-PN.

Design, setting, and participants

Consecutive cases of LESS-PN done between November 2007 and March 2012 at 11 participating institutions were included in this retrospective analysis.

Intervention

Each group performed LESS-PN according to its own protocols, entry criteria, and techniques.

Outcome measurements and statistical analysis

Demographic data, main perioperative outcome parameters, and perioperative complications were gathered and analyzed. A multivariable analysis was used to assess the factors predicting a short (≤20 min) warm ischemia time (WIT), the occurrence of postoperative complication of any grade, and a favorable outcome, arbitrarily defined as a combination of the following events: short WIT plus no perioperative complications plus negative surgical margins plus no conversion to open surgery or standard laparoscopy.

Results and limitations

A total of 190 cases were included in this analysis. Mean renal tumor size was 2.6, and PADUA score 7.2. Median operative time was 170 min, with median estimated blood loss (EBL) of 150 ml. A clampless technique was adopted in 70 cases (36.8%), and the median WIT was 16.5 min. PADUA score independently predicted length of WIT (low vs high score: odds ratio [OR]: 5.11 [95% confidence interval (CI), 1.50–17.41]; p = 0.009; intermediate vs high score: OR: 5.13 [95% CI, 1.56–16.88]; p = 0.007). The overall postoperative complication rate was 14.7%. The adoption of a robotic LESS technique versus conventional LESS (OR: 20.92 [95% CI, 2.66–164.64]; p = 0.003) and the occurrence of lower (≤250 ml) EBL (OR: 3.60 [95% CI, 1.35–9.56]; p = 0.010) were found to be independent predictors of no postoperative complications of any grade. A favorable outcome was obtained in 83 cases (43.68%). On multivariate analysis, the only predictive factor of a favorable outcome was the PADUA score (low vs high score: OR: 4.99 [95% CI, 1.98–12.59]; p < 0.001). Limitations of the study were the retrospective design and different selection criteria for the participating centers.

Conclusions

LESS-PN can be safely and effectively performed by experienced hands, given a high likelihood of a single additional port. Anatomic tumor characteristics as determined by the PADUA score are independent predictors of a favorable surgical outcome. Thus patients presenting tumors with low PADUA scores represent the best candidates for LESS-PN. The application of a robotic platform is likely to reduce the overall risk of postoperative complications.     

Meta-analysis of plasma to red blood cell ratios and mortality in massive blood transfusions for trauma

Background

The current military paradigm for blood transfusion in major trauma favours high plasma:RBC ratios. This study aimed determine whether high plasma:red blood cell (RBC) ratios during massive transfusion for trauma decrease mortality, using meta-analysis of contemporaneous groups matched for injury severity score.

Methods

A systemic review of the published literature for massive blood transfusions in trauma was performed. Patients were categorised into groups based on plasma:RBC transfusion ratios. Meta-analysis was only performed when there were no significant differences in Injury Severity Score (ISS) between ratio groups within studies. The main endpoint was 30-day mortality.

Results

Six observational studies reporting outcomes for 1885 patients were included in this meta-analysis. Five studies were from civilian environments and one from a military setting. Ratio cut-offs at 1:2 were the most commonly reported, demonstrating a survival advantage with higher ratios (OR 0.49, 95% CI 0.31–0.80, p = 0.004). Ratios ≥ 1:2 showed a significant reduction in mortality compared to lower ratios (OR 0.56, 95% CI 0.40–0.78, p < 0.001). Reducing the cut-off level was still protective (ratios between 1:2.5 and 1:4, OR 0.41), although the confidence interval was wide (0.16–1.00, p = 0.05) and data heterogenous (I² = 78%). Ratios of 1:1 were not proven to confer additional benefit beyond ratios of 1:2 (OR 0.50, 95% CI 0.37–0.68, p < 0.001).

Conclusions

In groups matched for ISS, there was a survival benefit with high plasma:RBC resuscitation ratios. No additional benefits of 1:1 over 1:2 ratios were identified.     

Prognostic factors for amputation in severe burn patients

Introduction

Amputation is a rare procedure among burned patients. However, it has significant physical and psychological consequences which impact quality of life.

Objective

To study the incidence, etiology and prognostic factors associated with amputation among burned patients in Chile.

Methods

Cohort study of patients admitted to the Reference Burn Center of Chile from 2006 to 2011. Association of demographic, event and injury variables with the likelihood of amputation were evaluated by using multivariable analysis.

Results

Amputation incidence was 5.8% in 1090 admitted patients. Male amputee patients were significantly more frequent (p = 0.01), with more electrical and high voltage burns (p < 0.01) and had greater frequency of impaired consciousness (p = 0.03). Multivariable analysis identified electrical burns (OR 13.7; 95% CI 6.7–28.1) and impaired consciousness (OR 2.8; 95% CI 1.4–5.7) as prognostic factors for amputation.

Conclusion

Amputation is a low incidence procedure among burned patients. Patients who underwent amputations are frequently at working age. Patients with high-voltage electrical burns and impaired consciousness are more likely to undergo amputation. Since these are highly incapacitating injuries, it is very important to implement preventive measures.     

Cemented hemiarthroplasty or hip replacement for intracapsular neck of femur fracture? A comparison of 7732 matched patients using national data

Background

The treatment of choice for intracapsular neck of femur (NOF) fractures in younger, more active patients remains unknown. Some surgeons advocate total hip replacement (THR).

Aim

This study aimed to compare complications following THR and hemiarthroplasty using the Hospital Episode Statistics (HES) database in England.

Method

Dislocation and revision rates were extracted for all patients with NOF fracture who underwent either cemented hemiarthroplasty or cemented THR between January 2005 and December 2008. To make a ‘like for like’ comparison all 3866 THR patients were matched to 3866 hemiarthroplasty patients (from a total of 41,343) in terms of age, sex and Charlson score.

Results and conclusion

Eighteen-month dislocation was significantly higher in the THR group (2.4% vs. 0.5%, odds ratio (OR) 3.90 (2.99–5.05), p < 0.001). This difference was sustained at the 4-year stage (2.9% vs. 0.9%, OR 3.18 (1.58–6.94), p = 0.001) in a subset of patients with longer follow-up. There was no significant difference in revision rate up to 4 years (1.8% vs. 2.1%, OR 0.85 (0.46–1.55), p = 0.666). In this national analysis of matched patients short- and medium-term dislocation rates following THR were significantly higher than following cemented hemiarthroplasty, without any difference in revision rates at 4 years. The low risk of dislocation may be acceptable in order to experience the apparent functional benefits of THR.     

Combination of CK20 and Ki-67 Immunostaining Analysis Predicts Recurrence,Progression, and Cancer-Specific Survival in pT1 Urothelial Bladder Cancer

Background

The prognostic value of CK20, Ki-67, and p53 has been investigated for non–muscle-invasive urothelial bladder cancers but not for the distinct and clinically challenging subset of pT1 bladder cancers.

Objective

To evaluate the prognostic value of CK20, Ki-67, and p53 within the largest series of pT1 urothelial bladder cancers.

Design, setting, and participants

Data from 309 patients with pT1 urothelial bladder cancer from one single urologic centre were collected.

Intervention

Adjuvant instillation of bacillus Calmette-Guérin was performed in each patient. A second resection was performed after 4–8 wk. A total of 76 patients underwent cystectomy.

Outcome measurements and statistical analysis

We conducted histomorphologic analysis; immunohistochemistry for CK20, Ki-67, and p53; and univariate and multivariate Cox regression models including recurrence-free survival (RFS), progression-free survival (PFS), and cancer-specific survival (CSS).

Results and limitations

At a median follow-up of 49 mo, we found recurrence and progression and disease-specific mortality rates of 22.7%, 20.1%, and 15.9%, respectively. CK20 expression was significantly correlated with RFS in multivariate analysis (hazard ratio [HR]: 5.89; 95% confidence interval [CI], 1.44–24.15; p = 0.014). In multivariate analysis, Ki-67 was the only marker significantly correlated with PFS (HR: 2.80; 95% CI, 1.45–5.43, p = 0.002). Ki-67 (HR: 3.83; 95% CI, 1.59–9.26; p = 0.003), and CK20 (HR: 8.44; 95% CI,1.16–61.34; p = 0.035) were significantly correlated with CSS in multivariate analysis. The combination of CK20 and Ki-67 showed significantly worse RFS (p = 0.026), PFS (p = 0.003), and CSS (p < 0.001) in tumours with a high proliferation index and abnormal CK20 expression. A retrospective study design was the major limitation of this study.

Conclusions

Our present analysis of the largest series of patients with pT1 urothelial bladder cancer published to date found Ki-67 and CK20 to be potential prognostic markers improving the risk stratification of pT1 bladder tumours. They are reliable indicators of biologic aggressiveness and may contribute to decision making on therapeutic strategy for pT1 bladder carcinomas.     

Features Associated with Recurrence Beyond 5 Years After Nephrectomy and Nephron-Sparing Surgery for Renal Cell Carcinoma: Development and Internal Validation of a Risk Model (PRELANE score) to Predict Late Recurrence Based on a Large Multicenter Database (CORONA/SATURN Project)

Background

Approximately 10–20% of recurrences in patients treated with nephrectomy for renal cell carcinoma (RCC) develop beyond 5 yr after surgery (late recurrence).

Objective

To determine features associated with late recurrence.

Design, setting, and participants

A total of 5009 patients from a multicenter database comprising 13 107 RCC patients treated surgically had a minimum recurrence-free survival of 60 mo (median follow-up [FU]: 105 mo [range: 78–135]); at last FU, 4699 were disease free (median FU: 103 mo [range: 78–134]), and 310 patients (6.2%) experienced disease recurrence (median FU: 120 mo [range: 93–149]).

Interventions

Patients underwent radical nephrectomy or nephron-sparing surgery.

Outcome measurements and statistical analysis

Multivariable regression analyses identified features associated with late recurrence. Cox regression analyses evaluated the association of features with cancer-specific mortality (CSM).

Results and limitations

Lymphovascular invasion (LVI) (odds ratio [OR]: 3.07; p < 0.001), Fuhrman grade 3–4 (OR: 1.60; p = 0.001), and pT stage >pT1 (OR: 2.28; p < 0.001) were significantly associated with late recurrence. Based on accordant regression coefficients, these parameters were weighted with point values (LVI: 2 points; Fuhrman grade 3–4: 1 point, pT stage >1: 2 points), and a risk score was developed for the prediction of late recurrences. The calculated values (0 points: late recurrence risk 3.1%; 1–3 points: 8.4%; 4–5 points: 22.1%) resulted in a good-, intermediate- and poor-prognosis group (area under the curve value for the model: 70%; 95% confidence interval, 67–73). Multivariable Cox regression analysis showed LVI (HR: 2.75; p < 0.001), pT stage (HR: 1.24; p < 0.001), Fuhrman grade (HR: 2.40; p < 0.001), age (HR: 1.01; p < 0.001), and gender (HR: 0.71; p = 0.027) to influence CSM significantly. Limitations are based on the multicenter and retrospective study design.

Conclusions

LVI, Fuhrman grade 3/4, and a tumor stage >pT1 are independent predictors of late recurrence after at least 5 yr from surgery in patients with RCC. We developed a risk score that allows for prognostic stratification and individualized aftercare of patients with regard to counseling, follow-up scheduling, and clinical trial design.     

The Application of Artificial Intelligence to Microarray Data: Identification of a Novel Gene Signature to Identify Bladder Cancer Progression

Background

New methods for identifying bladder cancer (BCa) progression are required. Gene expression microarrays can reveal insights into disease biology and identify novel biomarkers. However, these experiments produce large datasets that are difficult to interpret.

Objective

To develop a novel method of microarray analysis combining two forms of artificial intelligence (AI): neurofuzzy modelling (NFM) and artificial neural networks (ANN) and validate it in a BCa cohort.

Design, setting, and participants

We used AI and statistical analyses to identify progression-related genes in a microarray dataset (n = 66 tumours, n = 2800 genes). The AI-selected genes were then investigated in a second cohort (n = 262 tumours) using immunohistochemistry.

Measurements

We compared the accuracy of AI and statistical approaches to identify tumour progression.

Results and limitations

AI identified 11 progression-associated genes (odds ratio [OR]: 0.70; 95% confidence interval [CI], 0.56–0.87; p = 0.0004), and these were more discriminate than genes chosen using statistical analyses (OR: 1.24; 95% CI, 0.96–1.60; p = 0.09). The expression of six AI-selected genes (LIG3, FAS, KRT18, ICAM1, DSG2, and BRCA2) was determined using commercial antibodies and successfully identified tumour progression (concordance index: 0.66; log-rank test: p = 0.01). AI-selected genes were more discriminate than pathologic criteria at determining progression (Cox multivariate analysis: p = 0.01). Limitations include the use of statistical correlation to identify 200 genes for AI analysis and that we did not compare regression identified genes with immunohistochemistry.

Conclusions

AI and statistical analyses use different techniques of inference to determine gene–phenotype associations and identify distinct prognostic gene signatures that are equally valid. We have identified a prognostic gene signature whose members reflect a variety of carcinogenic pathways that could identify progression in non–muscle-invasive BCa.     

Procalcitonin in the Burn Unit and the Diagnosis of Infection

Objective

To determine the usefulness of procalcitonin (PCT) in decision-making when faced with suspected infection in patients with extensive burns.

Study

Retrospective, observational follow-up study.

Institution

Burn Unit of the Complexo Hospitalario Universitario A Coruña (CHUAC), Spain.

Patients and method

We included all patients admitted to the Unit from June 2011 to March 2012 with ≥20% total body surface area burned or ≥10% full-thickness body surface area burned with suspected infection (17 patients with 34 events of suspected infection).

Results

The infections were confirmed in 16/34 episodes (47.1%), and documented in 44.1% (n = 15). There were no statistically significant differences in the PCT figures at the time the infection was suspected between the cases with confirmed and unconfirmed infection (p = 0.682). The PCT values showed no discriminative value for differentiating patients with SIRS from those with sepsis, severe sepsis and septic shock (area under ROC curve (AUC) = 0.546; 95% CI: 0.326–0.766). No significant correlation was found between SOFA and PCT, although there were differences in the PCT values in the patients who had tissue hypoperfusion.

Conclusion

Results show that PCT is not a precise indicator of sepsis at the time of diagnosis. A correlation between PCT levels and hypoperfusion was observed.     

Impact de la mise en place d’un réseau de soins en traumatologie sur la mortalité des patients traumatisés graves du bassin

Aim

To evaluate the impact of a regional trauma network on intra-hospital mortality rates of patients admitted with severe pelvic trauma.

Study

Retrospective observational study.

Patients

Sixty-five trauma patients with serious pelvic fracture (pelvic abbreviated injury scale [AIS] score of 3 or more).

Methods

Demographic, physiologic and biological parameters were recorded. Observed mortality rates were compared to predicted mortality according to the Trauma Revised Injury Severity Score methodology adjusted by a case mix variation model.

Results

Twenty-nine patients were admitted in a level I trauma centre (reference centre) and 36 in level II trauma centres (centres with interventional radiology facility and/or neurosurgery). Patients from the level I trauma centre were more severely injured than those who were admitted at the level II trauma centres (Injury Severity Score [ISS]: 30 [13–75] vs 22 [9–59]; P < 0.01). Time from trauma to hospital admission was also longer in level I trauma centre (115 [50–290] min vs 90 [28–240] min, P < 0.01). Observed mortality rates (14%; 95% confidence interval, 95% CI, [1–26%]) were lower than the predicted mortality (29%; 95% CI [13–44%]) in the level I trauma centre. No difference in mortality rates was found in the level II trauma centres.

Conclusion

The regional trauma network could screen the most severely injured patients with pelvic trauma to admit them at a level I trauma centre. The observed mortality of these patients was lower than the predicted mortality despite increased time from trauma to admission.     

Photodynamic Diagnosis of Non–muscle-invasive Bladder Cancer with Hexaminolevulinate Cystoscopy: A Meta-analysis of Detection and Recurrence Based on Raw Data

Background

Studies on hexaminolevulinate (HAL) cystoscopy report improved detection of bladder tumours. However, recent meta-analyses report conflicting effects on recurrence.

Objective

To assess available clinical data for blue light (BL) HAL cystoscopy on the detection of Ta/T1 and carcinoma in situ (CIS) tumours, and on tumour recurrence.

Design, setting, and participants

This meta-analysis reviewed raw data from prospective studies on 1345 patients with known or suspected non–muscle-invasive bladder cancer (NMIBC).

Intervention

A single application of HAL cystoscopy was used as an adjunct to white light (WL) cystoscopy.

Outcome measurements and statistical analysis

We studied the detection of NMIBC (intention to treat [ITT]: n = 831; six studies) and recurrence (per protocol: n = 634; three studies) up to 1 yr. DerSimonian and Laird's random-effects model was used to obtain pooled relative risks (RRs) and associated 95% confidence intervals (CIs) for outcomes for detection.

Results and limitations

BL cystoscopy detected significantly more Ta tumours (14.7%; p < 0.001; odds ratio [OR]: 4.898; 95% CI, 1.937–12.390) and CIS lesions (40.8%; p < 0.001; OR: 12.372; 95% CI, 6.343–24.133) than WL. There were 24.9% patients with at least one additional Ta/T1 tumour seen with BL (p < 0.001), significant also in patients with primary (20.7%; p < 0.001) and recurrent cancer (27.7%; p < 0.001), and in patients at high risk (27.0%; p < 0.001) and intermediate risk (35.7%; p = 0.004). In 26.7% of patients, CIS was detected only by BL (p < 0.001) and was also significant in patients with primary (28.0%; p < 0.001) and recurrent cancer (25.0%; p < 0.001). Recurrence rates up to 12 mo were significantly lower overall with BL, 34.5% versus 45.4% (p = 0.006; RR: 0.761 [0.627–0.924]), and lower in patients with T1 or CIS (p = 0.052; RR: 0.696 [0.482–1.003]), Ta (p = 0.040; RR: 0.804 [0.653–0.991]), and in high-risk (p = 0.050) and low-risk (p = 0.029) subgroups. Some subgroups had too few patients to allow statistically meaningful analysis. Heterogeneity was minimised by the statistical analysis method used.

Conclusions

This meta-analysis confirms that HAL BL cystoscopy significantly improves the detection of bladder tumours leading to a reduction of recurrence at 9–12 mo. The benefit is independent of the level of risk and is evident in patients with Ta, T1, CIS, primary, and recurrent cancer.     

Pathway genetic load allows simultaneous evaluation of multiple genetic associations

Objective

Despite the general success of genome-wide association studies, much heritability remains unidentified in many disease states. Some of this ‘missing’ heritability may lie in epistatic interactions among multiple loci, which are typically ignored. We utilized a method for simultaneous evaluation of epistatic interactions between allelic variations within genes confined to a single pathway, which we have termed as pathway genetic load (PGL).

Methods

In separate analyses, we evaluated the risk for sepsis and for death associated with alleles at six loci in the TLR4 signaling and response pathway previously known or suspected to be linked to the development of sepsis after traumatic injury. We evaluated 155 patients with ≥15% TBSA burns and without significant non-burn trauma [ISS ≤ 16], traumatic or anoxic brain injury or spinal cord injury, who survived >48 h post-admission. Clinical data were collected prospectively and candidate genotypes were determined by TaqMan assay.

Results

After adjustment for burn size, inhalation injury, age, gender and race, PGL was associated with increased probability for complicated sepsis (aOR = 1.59; 95%CI = 1.11–2.29; p = 0.011) and death (aOR = 1.75; 95%CI = 1.11–2.76; p = 0.017).

Conclusion

Relative size and variability of aORs indicate greater power to detect genetic associations with PGL compared to the analysis of loci individually by multivariate logistic regression.     

Staphylococcus aureus bacteremia after thermal injury: The clinical impact of methicillin resistance

Objective

To evaluate the impact of methicillin resistance in Staphylococcus aureus bacteremia (SAB) on mortality and length of stay in burn patients.

Design

Retrospective cohort study.

Setting

A 750-bed tertiary care university hospital in Cologne, Germany.

Patients

Patients registered in the database of the burn intensive care unit (BICU) between 1989 and 2009 with complete data sets (n = 1688).

Results

Over the 21-year study period, 74 patients with SAB were identified; 33 patients had methicillin-resistant S. aureus (MRSA) and 41 methicillin-susceptible S. aureus (MSSA). Comparing the MRSA with the MSSA population the following parameters were significantly different in the univariate analysis: BMI (27.2 kg/m² vs. 23.6 kg/m²; P = 0.05), extent of deep partial thickness burns (17.8% vs. 9.0% of total body surface area; P = 0.007), antibiotic requirement on admission (45.5% vs. 22.0%; P = 0.046), median length of hospitalization prior SAB (24 days vs. 7 days; P < 0.001), packed red blood cells administration (47.6 units vs. 26.1 units; P = 0.003), intubation requirement (100% vs. 80.5%; P = 0.007), intubation period (43.5 days vs. 26.8 days; P = 0.008), catecholamine requirement (90.9% vs. 61.0%; P = 0.004), sepsis (60.6% vs. 34.1%; P = 0.035) and organ failures (81.8% vs. 39.0%; P < 0.001). Regarding outcome parameters, methicillin resistance was not significantly related with mortality (adjusted OR 1.55, 95% CI 0.56–4.28; P = 0.40) and length of BICU stay after SAB (Kaplan–Meier analysis log-rank test P = 0.32; Cox's proportional hazards regression HR 1.22, 95% CI 0.65–2.27, P = 0.535) in the univariate and multivariate analyses.

Conclusion

Our data suggest that methicillin resistance is not associated with significant increases in mortality and length of BICU stay among burn patients with SAB.