Risk Factors for Postoperative Urinary Tract Infections Following Anterior Lumbar Interbody Fusion

INTRODUCTION

Urinary tract infections (UTIs) are a common example of nosocomial infections in the postoperative period. Indwelling catheter use increases the risk for UTIs as the artificial plastic surface serves as a nidus for multiple species of bacterial growth that can lead to cystitis, and in some cases, pyelonephritis.^1,2 Accounting for almost 40% of all nosocomial infections, UTIs contribute significantly to increased length of stay, mortality risk, sepsis, superficial infection, and the economic burden following hospitalizations.^3–6 In the United States, many insurers have deemed nosocomial UTIs as preventable adverse events for which reimbursement will adversely affected, so there is an increasing financial pressure for surgeons and institutions alike to decrease the incidence of UTIs as well.⁷

Anterior lumbar interbody fusion (ALIF) has been a mainstay treatment modality for a variety of conditions affecting the lumbar spine, including degenerative disc disease, pseudarthrosis, and spondylolisthesis.^8–10 ALIF has been proposed to have several biomechanical advantages when compared to posterior approaches including preservation of posterior spinal musculature, more direct visualization of the disc space, and decreased development of adjacent segment disease.^11–13 Although risk factors contributing to UTI have been studied in posterior approaches to lumbar fusion, there is a lack of literature on factors contributing to UTI in ALIF.¹⁴ This current study seeks to establish risk factors that predispose patients undergoing ALIF to developing UTIs by utilizing The American College of Surgeons–National Surgical Quality Improvement Program (ASC-NSQIP) database so that surgeons can better stratify patients and take appropriate prophylactic measures to prevent certain postoperative complications that are associated with UTIs.

METHODS

The ACS maintains a multi-institutional clinical registry of over 5,000,000 surgical patients who have undergone procedures from 2005 to 2016 in its NSQIP database. The ACS-NSQIP database contains 274 variables for each surgical patient detailing demographics, operative parameters, and postoperative complications.

RESULTS

By querying the ACS-NSQIP database, we identified 10 232 patients who had undergone anterior approaches to the spine from 2005 to 2016, of whom 144 (1.41%) had developed postoperative UTI and 10 088 whom (98.59%) had not. These 2 cohorts of patients were analyzed for differences in various characteristics; those who had developed postoperative UTIs were significantly older (61.67 years, SD 13.283 versus 55.27 years, SD 13.627; P < .001) and more likely to be female (77.78% versus 53.37%; P < .001). These patients also had significantly higher levels of blood urea nitrogen (P = .011) and greater international normalised ratio (P = .003) but lower levels of serum albumin (P < .001) and hematocrit (P < .001) (Table 1).

There were significant differences in the prevalence of numerous preoperative comorbidities between the 2 cohorts. Patients who experienced UTIs had significantly higher rates of alcohol use (P = .043), dyspnea (P = .012), ventilator dependence (P = .008), hypertension requiring medications (P < .001), open wound/wound infections (P < .001), steroid use for chronic conditions (P = .007), significant weight loss (P = .014), and functional dependence (P = .013). UTI patients also had significantly higher rates of non–insulin-dependent diabetes mellitus but lower rates of insulin-dependent diabetes mellitus (P < .001), as well as smoking history (P = .007) (Table 1).

Patients who developed UTIs following ALIF had significantly longer operative times (250.19 minutes, SD 150.093 versus 198.16 min, SD 120.054; P < .001), total length of hospital stay (8.03 days, SD 8.103 versus 3.91 days, SD 4.856; P < .001), and inpatient stay following the operation (7.52 days, SD 7.290 versus 3.70 days, SD 3.367; P < .001). These patients also experienced increased discharge to destinations other than their own homes (P < .001), including separate acute care, skilled facilities, etc. (Table 2).

Upon multivariate analyses, several patient characteristics and comorbidities were found to be significant independent risk factors for postoperative UTIs in patients undergoing ALIF. Female patients were 3.1 times more likely to develop UTIs (95% confidence interval [CI] 2.08-4.64; P < .001) compared to men, and patients who were at least 60 years of age were 1.5 times more likely to experience postoperative UTIs (95% CI 1.07-2.20; P = .022) than younger patients. Several preoperative comorbidities were found to increase the risk for patients developing UTIs, including significant alcohol use (odds ratio [OR] 6.504, 95% CI 1.47-28.85; P = .014), open wounds or wound infections (OR 3.955, 95% CI 1.25-12.51; P = .019), and steroid use for chronic conditions (OR 1.965, 95% CI 1.01-3.82; P = .046). Longer operative times were also found to independently increase the risk for developing UTIs when compared to operations under 120 minutes, with ORs increasing for each subsequent time stratification: 120 min ≤ x < 180 min (OR 1.703, 95% CI 1.00-2.90; P = .050), 180 min ≤ x < 240 min (OR 1.895, 95% CI 1.09-3.31; P = .025), ≥ 240 min (OR 2.337, 95% CI 1.431-3.82; P = .001) (Table 3).

Patients who developed UTIs following their procedure also had significantly greater rates of other numerous postoperative complications. A complete list of postoperative complications and P values can be seen in Table 4. In addition, multivariate logistic regression models demonstrating increased risk for several complications in association with the UTI following ALIF can be found in Table 5.

DISCUSSION

The current retrospective study sought to identify risk factors that independently increased the risk for postoperative UTIs following ALIF. By identifying independent risk factors for UTI in patients undergoing ALIF, proper prophylactic measures can be taken to decrease the risk of UTI in these patients and prevent associated postoperative adverse events as UTIs have been associated with increased rates of postoperative complications, mortality, and increased length of stay.^3,4,6,17,18 Indeed, the results of this retrospective study contribute to this growing body of evidence as UTIs were associated with increased postoperative complications following ALIF (Table 5).

Risk factors identified in this study include age ≥ 60 years, female sex, alcohol use, chronic steroid use, preoperative open wound or wound infections, and longer operative times as independent risk factors for the development of UTIs following ALIF. To the authors' knowledge, there has been no previously reported association of chronic steroid use with UTI in ALIF. The results of this study support the current literature as the association between chronic steroid use and UTI occurrence has been previously reported in various urological procedures and total joint arthroplasty.^19,20 Steroids have been known to predispose patients to infectious complications as they are immunosuppressive via anti-inflammatory gene expression and alterations in immune cell trafficking.^21,22

In addition to chronic steroid use, this current study identifies significant alcohol intake as increasing the rate of UTI in ALIF patients. Alcohol has long been known to have deleterious effects via various cellular mechanisms when consumed in excess due to subclinical immunosuppressive properties that may only become apparent during physiological stress such as surgery.^23,24 Future research examining the efficacy of steroid tapering and alcohol cessation in decreasing UTI rates post-ALIF is warranted.

Female sex and age ≥ 60 years were also identified as risk factors for precipitating UTI in ALIF patients. Outside the setting of surgery, females have higher rates of UTIs compared to males when matched for age and other demographic variables due to (1) the shorter female urethra and (2) the proximity of the urethral opening to the end of the gastrointestinal tract.^1,25,26 Early genotypic studies of Escherichia coli in female UTIs have demonstrated identical genomic profiles to that of enteric E. coli.²⁷

Older age is another risk factor that increases the likelihood of developing a UTI following ALIF. Benign prostatic hypertrophy in older men leads to increased urinary stasis that promotes the growth of bacterium in the urinary tract.²⁸ Menopausal changes have also been known to predispose older women to UTI development due to decreases in estrogen in the vaginal epithelium.²⁹ Additionally, immune function declines as patients age, potentially preventing older individuals from mounting adequate immune responses to prevent UTI.³⁰ Bohl et al previously reported that female sex increased the risk for UTI in posterior approaches to lumbar fusion. In addition, patients ≥ 60 years old had an increased risk of UTI that correlated positively with increasing age, whereas patients ages 50 to 59 years old had no difference in UTI rates compared to a younger population cohort.¹⁴ As age and sex are nonmodifiable risk factors for UTI, a lower threshold of suspicion for UTI should be held for patients undergoing ALIF who are female and ≥ 60 years of age. Although appropriate removal of catheters is a standard of care,²⁵ stringent adherence to guidelines for these high-risk patients is warranted to prevent further morbidity.

The current retrospective study lends further supporting evidence to a growing body of literature that reports longer operative times lead to increased UTI development. We report operative times greater than 119 minutes demonstrate a direct positive correlation with increased postoperative UTI development—each additional stratification of operative time demonstrated increased statistical significance. Previous reports have shown that longer operative times increase the risk of UTI across all surgical fields, including posterior lumbar fusion and single-level lumbar fusion of different approaches.^14,31–33 To the authors' knowledge, this relationship has not been previously demonstrated in anterior approaches to the lumbar spine. The ACS-NSQIP database does not report why certain operations are of longer duration or the specific length of operative times, likely representing limitations to this study. Longer operative times can be due to a number of factors including additional levels, excessive bleeding requiring operative control, high soft-tissue body habitus, and surgeon experience or case volume. Longer operative times have been reported to increase the risk of pathogen exposure and increased iatrogenic tissue trauma that may increase the risk of UTI.³² Increased operative time can also lead to UTI via hypothermia induced lymphocytic dysfunction.³⁴ In addition, an increased operative time could also indicate a more complex operation that could require extended postoperative catheterization, allowing bacteria more time to ascend the urinary tract and cause infection.¹ Further quality improvement focused work investigating methods to reduce operative time in ALIF may prove beneficial in reducing UTI rates.

Our study identified preoperative open wounds or wound infections to be an independent risk factor for the development of UTI in ALIF patients. UTIs have been reported to increase the risk of various infectious complications in the postoperative period following many surgical procedures. Asymptomatic and symptomatic preoperative UTIs and postoperative UTIs have been linked to increased rates of infectious complications, sepsis, morbidity, and mortality.^35,36 Multivariate analyses in this study provides supporting evidence as UTIs were found to increase the risk for postoperative surgical site infection and pneumonia as well. These complications likely arise from more severe UTIs that can lead to urosepsis, widespread migration of bacteria throughout the body, and subsequent infectious complications. However, to the authors' knowledge, there appears to be no other reports of the inverse relationship.

There is a lack of literature investigating preoperative wounds and their association with UTI in the postoperative period. The exact link between preoperative wounds that lead to UTI in the postoperative period is as of yet unclear. It is plausible that those with preoperative wound infections may already be immunocompromised, leading to an increased likelihood of developing UTIs following ALIF. As the ACS-NSQIP database does not provide specific parameters of immune functioning in these patient files, this represents a limitation of the database for this study. Staphylococcus aureus has been implicated in primary S. aureus UTI as the bacterium can seed the urinary tract hematogenously as opposed to fecal-perineal-urethral contamination by enteric organisms.³⁷ Hematogenous spread of staphylococcal species may very well lead to UTI in ALIF patients. Future work analyzing the correlation of nonenteric bacterium in UTI and the presence of preoperative open wounds or wound infections is warranted. In light of these results, surgeons should postpone elective ALIF operation until all open wounds and wound infections have been resolved to reduce the risk of UTI and subsequent increased morbidity.

This current retrospective study presents with several limitations. This study only analyzed demographic variables, comorbidities, and complications in which ≥ 85% of the patient files were recorded. By excluding other variables that did not meet these inclusion criteria, several other demographic factors, postoperative complications, and comorbidities that may in reality be associated with UTIs may have not been reported. In addition, the inability to verify the information being entered into the database is a limitation inherent to all nationwide database studies. Furthermore, CPT codes are not designed for research purposes—the use of CPT codes can introduce coding bias based on financial reimbursement.³⁸ This current study included all ALIF procedure files that met inclusion criteria. However, as ALIF is indicated for a variety of spinal conditions, distinct pathologies may represent systematic confounding variables that this retrospective study could not account for. An additional confounding variable not accounted for in this study is that all ALIF procedures, including multilevel operations, were included. Future studies examining the effect of multilevel operations on postoperative UTI rates may prove to demonstrate increasing risk of UTI with increasing levels of operation. Finally, the ACS-NSQIP database does not provide details on catheter use. Specific information regarding duration of catheterization in ALIF patients could provide data on whether there is a particular period of time that UTI risk becomes unacceptable in the postoperative period. Elucidating a specific catheterization duration after which the risk of UTI becomes unacceptable may prove useful in formulating quantitative guidelines in the future.

CONCLUSIONS

The present study demonstrated that age ≥ 60 years, female sex, alcohol use, chronic steroid use, preoperative open wound/wound infections, and longer operative times increase the risk for postoperative UTI following ALIF. Identification of risk factors for UTI after ALIF may allow surgical care teams to better risk stratify patients in order to initiate more aggressive prophylactic measures to mitigate the effects of UTIs and their associated complications. The earliest possible removal of catheters must be stringently adhered to in these patients. Providers should have lower thresholds of suspicion for high-risk patients in the postoperative period following ALIF to avoid UTI and its associated complications. Quality improvement initiatives focusing on ensuring earliest postoperative catheter removal may help mitigate risk in these high-risk patients.