International Journal of Surgery 20 (2015) 107e112

Contents lists available at ScienceDirect

International Journal of Surgery journal homepage: www.journal-surgery.net

Original research

Surgical duration and risk of Urinary Tract Infection: An analysis of 1,452,369 patients using the National Surgical Quality Improvement Program (NSQIP) Charles Qin a, Gildasio de Oliveira b, Nicholas Hackett a, John Y.S. Kim a, * a b

Department of Surgery, Northwestern University, Feinberg School of Medicine, Chicago, IL, USA Department of Anesthesiology, Northwestern University, Feinberg School of Medicine, Chicago, IL, USA

h i g h l i g h t s  We have evaluated the relationship between surgical duration and risk of 30-day Urinary Tract Infections (UTI).  Inter-procedural differences in surgical duration were standardized with a z-score and quintile stratification system.  Upon regression analysis, longer surgical durations confer increased risk of UTI across all surgeries and specialties.  This trend persisted in multiple sub-analyses.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 10 March 2015 Received in revised form 27 May 2015 Accepted 31 May 2015 Available online 6 June 2015

Introduction: While the relationship between surgical duration and post-operative morbidity has been well-studied in specific procedures for specific complications, there is a paucity of literature that addresses whether longer surgeries increase the risk of Urinary Tract Infection (UTI). We have performed the first study to elucidate the relationship between increasing surgical duration and UTI events across surgical specialties via the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database. Methods: Patients, who received general anesthesia, were stratified into quintiles by a calculated z-score for their anesthesia time based on the standard deviation and mean of their respective current procedural terminology (CPT) code. Z-score analysis standardized interprocedural differences in anesthesia time. Multivariate regression analysis was employed to evaluate the independent association of anesthesia time with risk of UTI. Multiple sub-analyses were performed to evaluate the robustness of our results. Results: 22,305 patients (1.5%) experienced a UTI. Compared to the mean procedural duration as represented by the 3rd quintile, procedures of longer duration were independently associated with increased risk of UTI (OR, 1.156 (95% CI 1.104e1.21); OR, 1.758 (95% CI 1.682e1.838)) while procedures of shorter duration were associated with reduced risk (OR, .928 (95% CI .873-.987); OR, .955 (95% CI .906 e1.007)). Conclusions: Our findings suggest that increasing surgical duration may independently worsen the risk of post-operative UTI pan-surgically. We hope that our results will help guide decision making regarding the safety of combination procedures as well as improve pre-operative risk stratification. Published by Elsevier Ltd on behalf of IJS Publishing Group Limited.

Keywords: Surgical duration Anesthesia time Post-operative UTI 30-Day outcomes NSQIP

1. Introduction * Corresponding author. Division of Plastic and Reconstructive Surgery, Northwestern University, Feinberg School of Medicine, 675 North St. Clair Street, Galter Suite 19-250, Chicago, IL 60611, USA. E-mail addresses: [email protected] (C. Qin), [email protected] (G. de Oliveira), [email protected] (N. Hackett), [email protected] (J.Y.S. Kim). http://dx.doi.org/10.1016/j.ijsu.2015.05.051 1743-9191/Published by Elsevier Ltd on behalf of IJS Publishing Group Limited.

Urinary tract infections (UTI) are a well-documented burden for operative patients, representing 40% of all hospital-acquired infections [1,2]. Given that UTI has been linked to higher hospital costs, increased length of stay, and increased risk for mortality, initiatives have been developed in an effort to reduce the incidence

108

C. Qin et al. / International Journal of Surgery 20 (2015) 107e112

of UTI's [1,3]. Most notably, the Centers for Medicare and Medicaid Services (CMS) have withdrawn additional payment for UTI events. While time will tell whether such initiatives will be effective, reports of increasing annual rates of UTI's are not promising [1]. Identifying risk factors associated with UTI events in an effort to enable surgeons to risk stratify patients, counsel patients on modifiable risk factors, and make safer outcomes-based decisions is an active area of investigation. The relationship between surgical duration and post-operative morbidity has been well-studied in specific procedures for specific complications, but there is a paucity of literature that addresses whether longer surgeries increase the risk of UTI [4e8]. While the duration of catheterization remains the most widely reported risk factor for UTI, it is unknown if surgical duration independently confer increased risk of sustaining UTI in all procedures [9e12]. If proven to be independently associated, our findings may guide decision making regarding the safety of combination procedures and the efficacy of operative techniques and materials that shorten procedure times. The main objective of the current investigation was to evaluate a relationship between surgical duration and the development of post-surgical UTI. We hypothesized that surgical duration would independently confer increased risk of sustained UTI in all procedures captured by the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP). 2. Methods 2.1. Data source To evaluate this hypothesis from a national perspective, we analyzed data from the American College of Surgeon's National Surgical Quality Improvement Program (NSQIP) database. The NSQIP was instituted in 2004 and provides comprehensive information collected by trained surgical nurses for major surgical procedures at more than 300 institutions across the United States, on over 240 patient variables [13]. It plays a crucial role in continued quality improvement in the surgical field [14]. Via this database from which we queried over 1.4 million patients, we performed the first multi-institutional analysis to evaluate whether increasing surgical duration, defined as total anesthesia time, was independently associated with an increase in the incidence of UTI. Reliability of the data and collection methods has previously been described. The Northwestern University IRB determined that this study does not constitute research with human subjects and was thus exempt from formal IRB approval. 2.2. Patient population and variables of interest All general anesthesia cases reported between 2005 and 2012 in NSQIP were included in our study. All cases using other anesthesia types were excluded. Surgical duration was measured by the variable that captured total duration of general anesthesia as defined by NSQIP. NSQIP does not capture the type of anesthetic used for maintenance. All cases with a missing anesthesia time were excluded. Given the heterogeneous distribution of anesthesia times among different procedures, we ensured standardization of surgical duration by calculating a z-score for each case. The difference between individual anesthesia time and the mean for its respective current procedural terminology (CPT) code was divided by its standard deviation. Cases were divided into 5 equal quintiles based on this z-score with the 1st quintile representing procedures with shortest surgical duration, the 3rd quintile representing procedures of average duration, and the 5th quintile representing the longest procedures. The outcome of interest was Urinary Tract Infection (UTI). A UTI was diagnosed based on classic symptoms of urgency,

fever, frequency, dysuria, suprapubic tenderness and a positive urine culture. All patients with pre-operative systemic infection were excluded from our analysis. 2.3. Statistical analysis Patient demographic data queried for univariate comparison and risk adjustment included age, body mass index (BMI), gender, and race. Clinical characteristics and comorbidities included sex, inpatient status, diabetes mellitus (DM), smoker, acute renal failure, wound infection, pre-operative transfusion, ASA Class, race, surgical specialty, age, BMI, WBC count, and total RVU. Surgical specialties included in our analysis included general, cardiac, orthopedic, vascular, otolaryngology, plastic, thoracic, urologic, gynecologic, and neurosurgery. Because the relationship of catheter placement duration on catheter-associated UTIs (CAUTI) in the inpatient settings is well-recognized, length of surgical stay was included as a surrogate in our risk-adjusted model, the implications of which are discussed later. Variables were selected with clinical judgment. Categorical variables were analyzed using chi-square test and continuous variables were analyzed using one-way ANOVA tests. To evaluate the independent effect of surgical duration on UTI rates, we performed a multivariate logistic regression analysis including the above covariates. Our regression model quantified the risk that relatively short (1st and 2nd quintiles) and relatively long (4th and 5th quintiles) procedures confer relative to procedures of relative length (3rd quintile). The influence of the z-score of anesthesia time as a continuous variable was also examined with separate analyses. The robustness and generalizability of our results was tested with subsequent analyses. They included subgroup analysis for surgical specialties and admission type. In our sub-group analysis by surgical specialty, odds ratios were generated for both by hour increase in surgical duration as well as by standard deviations from the mean. The effect of outliers was accounted for via an analysis that eliminated outliers defined as a z-score greater than 3 or less than 3. HosmereLemeshow (H-L) and c-statistics were calculated to assess model calibration and discriminatory capability, respectively [15,16]. Significance for the H-L statistic was set at P < .05. All analysis was performed using SPSS version 22 (IBM Corp Armonk, NY). 3. Results A total of 2,320,920 surgical cases were initially identified in the NSQIP database from 2005 to 2012. After excluding patients who did not receive general anesthesia or received general anesthesia for an unspecified duration, and patients with pre-operative sepsis, 1,452,615 patients remained and were included in our analysis. Patients were stratified by quintiles based on the z-score of anesthesia time calculated. Comparison of demographic profiles and clinical characteristics were made across the quintile groups (Table 1). All pre-operative and demographic factors were significantly different. 22,305 (1.5%) patients were reported to have a UTI. Absolute rates of UTI were calculated for each quintile group (Supplementary Table 2). A stepwise increase in UTI rates was observed across quintiles. Multivariate regression analysis was employed to determine the risk of UTI in procedures of shorter and longer surgical durations relative to procedures of average duration, represented by the 3rd quintile (Table 2). The absolute risk of UTI occurrence for the 3rd quintile was 1.40%. Relative to the 3rd quintile, the 1st and 2nd quintiles carried adjusted odds ratios (OR) of .93 and .96, respectively. The 4th and 5th quintiles carried OR of 1.16 and 1.76. When

C. Qin et al. / International Journal of Surgery 20 (2015) 107e112

109

Table 1 Pre-operative characteristics and demographics across surgical time groupings.

Male Inpatient Diabetes Smoker Acute renal failure Wound infection Pre-operative transfusion ASA class 3,4,5 Race White Asian Black Other Not Available Emergency case Age BMI White blood cell count RVU

1st Quintile

2nd Quintile

3rd Quintile

4th Quintile

5th Quintile

P value

38.40% 57.60% 13.40% 20.40% 0.20% 2.90% 0.50% 40.50% 76.00% 1.70% 7.00% 11.00% 4.30% 7.8% 55.9 (17.1) 29.2 (8.1) 7.7 (3.0) 20.8 (14.4)

40.60% 58.80% 13.30% 20.00% 0.20% 3.20% 0.40% 40.00% 75.80% 2.10% 8.80% 10.30% 2.90% 13.8% 54.9 (17.2) 29.5 (8.1) 7.9 (3.2) 20.0 (14.9)

42.10% 62.40% 14.10% 19.70% 0.30% 3.40% 0.50% 42.00% 75.50% 2.20% 9.90% 10.00% 2.40% 11.7% 54.9 (17.0) 29.9 (8.3) 7.9 (3.2) 21.2 (16.3)

43.40% 67.10% 15.00% 19.70% 0.30% 3.30% 0.50% 44.50% 74.50% 2.20% 11.10% 10.00% 2.20% 5.6% 55.0 (16.6) 30.4 (8.6) 7.8 (3.1) 23.3 (18.4)

45.00% 75.10% 16.50% 19.80% 0.30% 3.70% 0.60% 49.30% 73.80% 2.10% 12.20% 9.80% 2.10% 4.2% 55.3 (15.9) 31.2 (19.9) 7.7 (3.0) 28.6 (23.9)

Surgical duration and risk of Urinary Tract Infection: An analysis of 1,452,369 patients using the National Surgical Quality Improvement Program (NSQIP).

While the relationship between surgical duration and post-operative morbidity has been well-studied in specific procedures for specific complications,...
216KB Sizes 3 Downloads 7 Views

Recommend Documents