Antibiotic prophylaxis for subdural and subgaleal drains

Transcription

1 CLINICAL ARTICLE J Neurosurg 126: , 2017 Antibiotic prophylaxis for subdural and subgaleal drains Ariane Lewis, MD, 1,2 Rajeev Sen, BA, 3 Travis C. Hill, PhD, 2,3 Herbert James, BS, 3 Jessica Lin, MD, 1 Harpaul Bhamra, MD, 1 Nina Martirosyan, BS, PharmD, 4 and Donato Pacione, MD 2 Departments of 1 Neurology and 2 Neurosurgery; 3 NYU School of Medicine; and 4 Pharmacy, NYU Langone Medical Center, New York, New York OBJECTIVE The authors sought to determine the effects of eliminating the use of prolonged prophylactic systemic antibiotics (PPSAs) in patients with subdural and subgaleal drains. METHODS Using a retrospective database, the authors collected data for patients over the age of 17 years who had undergone cranial surgery at their institution between December 2013 and July 2014 (PPSAs period) or between December 2014 and July 2015 (non-ppsas period) and had subdural or subgaleal drains left in place postoperatively. RESULTS One hundred five patients in the PPSAs period and 80 in the non-ppsas period were identified. The discontinuation of PPSAs did not result in an increase in the frequency of surgical site infection (SSI). The frequency of Clostridium difficile (CDI) and the growth of resistant bacteria were reduced in the non-ppsas period in comparison with the PPSAs period. In the 8 months after the drain prophylaxis protocol was changed, $93, were saved in the costs of antibiotics and complications related to antibiotics. CONCLUSIONS After discontinuing PPSAs for patients with subdural or subgaleal drains at their institution, the authors did not observe an increase in the frequency of SSI. They did, however, note a decrease in the frequency of CDI and the growth of resistant organisms. It appears that not only can patients in this population do without PPSAs, but also that complications are avoided when antibiotic use is limited to 24 hours after surgery. KEY WORDS antibiotics; drains; surgical site infection; prophylaxis; postoperative infection Surgical site infections (SSIs) increase postoperative morbidity and mortality rates and lead to prolonged hospitalizations. As a result, neurosurgeons are constantly striving to determine the ideal tactics for avoiding postoperative infection. 2 One such method is the administration of prolonged prophylactic systemic antibiotics (PPSAs) to patients with neurosurgical drains and devices. 10,20,21 Although the goal of administering PPSAs is to protect a patient from complications, PPSAs have been associated with the development of nosocomial infections such as Clostridium difficile (CDI) 9,20 and the growth of resistant bacteria. 5,11,19 At our institution, the antibiotic prophylaxis protocol for patients with subdural and subgaleal drains was recently changed so that we no longer administer PPSAs to this patient group. In this setting, we sought to retrospectively evaluate the risks and benefits of PPSAs in this population. Our primary objective was to determine whether the discontinuation of PPSAs was associated with an increase in SSIs. Our secondary objectives were to determine if PPSAs discontinuation was associated with any change in the rate of CDI or infection with resistant organisms. Lastly, we assessed the change in costs associated with the discontinuation of PPSAs. Methods Antibiotic Prophylaxis Protocol Prior to October 2014, every patient who underwent a neurosurgical procedure at NYU Langone Medical Center and had a subdural or subgaleal drain left in place postoperatively was treated with PPSAs (cefazolin or vancomycin if cefazolin was contraindicated) until the drain was removed. In October 2014, the Neurosurgery Quality Improvement Committee changed our institutional protocol so that all patients received at least 1 dose of antibiotics intraoperatively within 1 hour of incision, but the postoperative administration of prophylactic antibiotics was limited to a maximum of 24 hours after surgery. No other systematic changes to infection prevention were made during this time period, and the surgeons who operated before and after the policy change remained the same. ABBREVIATIONS CDC = Centers for Disease Control and Prevention; CDI = Clostridium difficile; PPSAs = prolonged prophylactic systemic antibiotics; SSI = surgical site infection. SUBMITTED February 2, ACCEPTED April 1, INCLUDE WHEN CITING Published online June 3, 2016; DOI: / JNS AANS, 2017

2 Antibiotic prophylaxis Patient Selection and Data Collection Data were collected from a retrospective quality-improvement database for patients over the age of 17 years who had undergone cranial surgeries and had had subdural or subgaleal drains placed at our institution in the PPSAs period (December 2013 to July 2014) or the non- PPSAs period (December 2014 to July 2015). These dates were selected to allow for a washout period just before and after the time of the protocol change. Patients with hardware (for example, subdural grids or a ventriculoperitoneal shunt) or any other type of drain (for example, an external ventricular drain, a lumbar drain, intranasal packing for endonasal procedures, or an abdominal drain) were excluded because these patients were not covered under the scope of the antibiotic prophylaxis protocol changes at our institution. Patients with known intracranial infections (for example, an abscess or meningitis) prior to surgery were also excluded as they required prolonged antibiotics for treatment rather than for prophylaxis. Patients who underwent surgery from December 2013 to July 2014 and were not given a full course of PPSAs (defined as drain days ± 1) and those who underwent surgery from December 2014 to July 2015 and were given PPSAs for more than 1 day were excluded as their treatments represented protocol violations. Additionally, patients who received prophylactic antibiotics other than cefazolin or vancomycin were excluded. Of note, our record of antibiotic doses reflects postoperative antibiotics only and does not include doses administered intraoperatively. Our database included culture data for 90 days after drain placement. Deep and superficial SSIs were defined using criteria set forth by the Centers for Disease Control and Prevention (CDC). A deep SSI is present if a patient has one or more of the following: 1) purulent drainage from a deep incision, 2) an incision that dehisces or is intentionally debrided and is culture positive or not cultured and is associated with fever > 38 C or localized pain with a positive culture or with no cultures sent, or 3) an abscess at the surgical site. A superficial SSI is present if a patient has one or more of the following: 1) purulent drainage from a superficial incision; 2) organisms isolated from a culture of a superficial incision; or 3) an incision that was debrided and is culture positive or not cultured and is associated with pain, swelling, redness, or heat. 7 Growth of resistant organisms was defined as growth of an organism that was resistant to the intravenous antibiotic used for prophylaxis (cefazolin or vancomycin). If no systemic antibiotic was used postoperatively for prophylaxis, we broadly defined a resistant organism as one resistant to cefazolin or vancomycin. Cost Calculations Cost savings for antibiotics were calculated based on the cost of a 1-g bag of cefazolin ($3.29) and a 1-g vial of vancomycin ($3.10) at our institution. We assigned a cost of $7286 to CDI, as calculated by Magee et al. after a review of 171,586 patients in the Premier health care database. 18 For resistant infections, we assigned a cost of $25,573 based on calculations by Roberts et al. after a review of an electronic database of patients at 4 hospitals in New York (the location of our institution). 22 Data Evaluation Data were analyzed using descriptive statistics and Fisher, chi-square, and t-tests, as appropriate. All statistical analyses were performed using SPSS Statistics 21 (IBM Corp.). Sample size calculations were performed using MATLAB 2015b. A p value of 0.05 was considered statistically significant. The institutional review board at NYU Langone Medical Center approved this study. Results One hundred five patients in the period with PPSAs and 80 in the period without PPSAs met our inclusion criteria (Table 1). Six patients were excluded due to protocol violations (1 in the PPSAs period and 5 in the non-ppsas period). Violations appeared to be arbitrary, and there was no evidence that these patients were at higher or lower risk for infection (Table 2). During the PPSAs period, a total of 513 doses of cefazolin and 77 doses of vancomycin were administered for drain prophylaxis. In the non-ppsas period, only 6 doses of cefazolin and 1 dose of vancomycin were administered for drain prophylaxis. Table 3 summarizes the infection data for the PPSAs and non-ppsas periods. There was no change in the frequency of SSI after the discontinuation of PPSAs. During the PPSAs period, there were 3 patients who grew resistant organisms. After PPSAs was discontinued, no patients grew resistant organisms. There were 2 cases of CDI during the PPSAs period, but no cases after the discontinuation of PPSAs. Table 4 has the details on the patients who developed SSI or CDI or grew resistant organisms. The mean length of stay for the 4 patients who grew resistant organisms or had CDI was 17 days (SD 14 days). The mean TABLE 1. Demographic data in 185 patients with subdural or subgaleal drains after cranial surgery Patients in PPSAs Period Non-PPSAs Period p Value* No. of patients No. of males (%) 58 (55) 32 (40) 0.05 Mean age in yrs (SD) 59 (17) 57 (17) 0.36 Mean no. of hospital days (SD) 5 (4) 5 (4) 0.6 Primary diagnosis (no. [%]) Hemorrhage 30 (29) 5 (6) Tumor-benign 32 (30) 33 (41) Tumor-malignant 14 (13) 19 (24) Vascular lesion 27 (26) 18 (22) Other 2 (2) 5 (6) Drain data Mean no. of drains (SD) 1 (0.3) 1 (0.2) 0.64 Mean no. of drain days (SD) 3 (0.7) 3 (0.9) 0.84 Prophylaxis Total cefazolin doses Total vancomycin doses 77 1 SD = standard deviation. * Boldface type indicates statistical significance. 909

3 A. Lewis et al. TABLE 2. Demographic data for patients with protocol violations Patient No Study period PPSAs Non-PPSAs Non-PPSAs Non-PPSAs Non-PPSAs Non-PPSAs Sex F F F M M M Age in yrs No. of hospital days Primary diagnosis Vascular lesion Vascular lesion Tumor-malignant Hemorrhage Tumor-benign Hemorrhage No. of drains No. of drain days No. of antibiotic days length of stay for the remaining 181 patients in both the PPSAs and non-ppsas periods was 5 days (SD 3 days). Based on our assumptions, the cost of antibiotics and complications during the PPSAs period was $93, ($ in direct costs for antibiotics; $14,572 in costs related to CDI; and $76,719 in costs attributed to resistant infections), and the cost in the non-ppsas period was $22.84 (for direct antibiotic costs). The cost of antibiotics and complications per patient was $ in the PPSAs period and $0.29 in the non-ppsas period. Thus, in the 8 months after the drain prophylaxis protocol was changed, we spent a total of $93, less than in the PPSAs period ($ less per patient). Discussion Postoperative wound infections affect 920,000 of the 23 million patients who undergo surgery every year. They are the most expensive type of hospital-acquired infection and can impact patient morbidity and mortality. As a result, surgeons strive to prevent SSI by optimizing preoperative patient preparation, surgical technique, perioperative antibiotic prophylaxis, and postoperative wound care. 3 Infections after cranial surgery most commonly present as meningitis, cerebral abscess, or subdural empyema. 8 Placement of a foreign body, such as a drain, can increase the risk of postoperative infection. 6 The administration of antibiotic prophylaxis to patients after cranial surgery varies widely between institutions. In a large retrospective study, Dashti et al. reviewed more than 16,000 cranial surgeries at their institution and described a lack of uniformity in antibiotic prophylaxis. 8 While some patients did not receive any antibiotics, the majority received a 24-hour postoperative course of antibiotics regardless of the type or duration of surgery. Korinek et al. described variations in antibiotic administration based on the level of contamination of the surgical site (clean vs clean-contaminated), type of procedure (elective vs emergency), and duration of procedure. 14 There was no mention of protocol modifications for postoperative drains. Contrastingly, Akiyama et al. reported the administration of antibiotics for 1 week after bur hole drainage for subdural hematomas. 4 TABLE 3. Summary of infection data PPSAs Period Non-PPSAs Period OR (95% CI) p Value No. of patients SSIs No. of deep infections (%) 1 (1) 0 (0) 1 (0.99 1) 1 No. of superficial infections (%) 1 (1) 1 (1) 0.77 ( ) 1 Blood Growth of blood culture (no. [%]) 3 (3) 0 (0) 2.1 (0.9 5) 0.1 Growth of resistant organism in blood (no. [%]) 1 (1) 0 (0) 1 (0.99 1) 1 CSF Growth of CSF culture (no. [%]) 0 (0) 0 (0) Growth of resistant organism in CSF (no. [%]) 0 (0) 0 (0) Sputum Growth of sputum culture (no. [%]) 3 (3) 2 (2) 1.3 ( ) 1 Growth of resistant organism in sputum (no. [%]) 1 (1) 0 (0) 1 (0.99 1) 1 Urine Growth of urine culture (no. [%]) 12 (11) 9 (11) 1.5 ( ) 0.26 Growth of resistant organism in urine (no. [%]) 1 (1) 0 (0) 1 (0.99 1) 1 CDI 2 (2) 0 (0) 1 (0.99 1)

4 Antibiotic prophylaxis TABLE 4. Characteristics of patients with SSI, CDI, or resistant organisms within 90 days of drain placement Infection PPSA Age (yrs) Sex Primary Diagnosis Antibiotic Prophylaxis No. of Days of Prophylaxis No. of Days Btwn Drain Placement & Infection Deep SSI: Propionibacterium acnes Yes 24 M Tumor-benign Cefazolin 3 34 Superficial SSI: Klebsiella Yes 62 M Hemorrhage Cefazolin 3 51 Superficial SSI: no organism identified No 47 F Tumor-benign None 0 85 Resistant Bacillus in blood & CDI Yes 33 M Tumor-benign Cefazolin 3 16 & 22 Resistant Serratia in sputum Yes 76 M Hemorrhage Cefazolin/vancomycin* 4 10 Resistant Enterobacter in urine Yes 51 F Vascular lesion Cefazolin 3 10 CDI Yes 90 F Hemorrhage Cefazolin 2 3 * This patient was initially given cefazolin for prophylaxis but treatment was then changed to vancomycin. Monitoring outcomes to evaluate risks and benefits of PPSAs administration allows for the development of evidence-based guidelines. 3 Our experience suggests that patients with subdural or subgaleal drains do not require PPSAs to prevent SSI and, in fact, that the use of PPSAs in this population is associated with an increased risk of CDI and growth of resistant bacteria. Overall, our frequencies of infection in both the PPSAs and non-ppsas periods were better than those for patients undergoing craniotomy as reported by Kourbeti et al., who noted that more than 40% of patients developed at least one infection postoperatively, most commonly ventilatorassociated pneumonia. 15 These authors noted that the use of antibiotic prophylaxis did not affect the rate of meningitis development, which was in line with our results. The vast difference in the infection rate between their study and ours reflects the difference in patient populations, as the majority of our patients underwent elective procedures and the majority of their patients presented after trauma. Accordingly, the median length of stay in their cohort was 27 days in comparison with a mean of 5 days in both of our patient groups. Our rates of SSI and CDI are similar to those reported in prior studies on postoperative infection. 1,2 Although it can be tempting to administer antibiotics to patients to be safe 17 or just in case, 13 our results show that PPSAs use is not associated with a protective benefit, but instead is associated with a risk of harm. Our findings are consistent with the observations of Abu Hamdeh et al., who reported no significant relationship between the presence of a drain and the development of an SSI in their prospective study of 448 patients undergoing intracranial tumor resections, bur hole biopsies, or bur hole placement for subdural hematoma drainage. 2 Of note, patients in that study only received preoperative antibiotic prophylaxis. Thus, our work adds to a growing body of literature that suggests that the risks of PPSAs outweigh the benefits for a subset of neurosurgical procedures. Aside from the significant costs associated with CDI and resistant infections, 18,22 these complications have a substantial impact on patient lives. Clostridium difficile infection increases patient mortality and length of stay in the hospital. In some cases, it can lead to fulminant colitis and require surgical intervention. 16 Resistant infections are similarly associated with delayed recovery, recurrent infections, and increased risk of death. 22 Both of these complications impact not just a single patient receiving PPSAs, but also all patients in the hospital because antibiotic use affects the flora to which all patients are exposed. 3,19 Because antibiotic use is associated with the risk of nosocomial infections and growth of resistant bacteria, 5,9,10,19,20 the CDC is exploring approaches to improve monitoring and reporting of inpatient antimicrobial use. 11 The Joint Commission monitors the use of prophylactic antibiotics and rates of discontinuation after 24 hours for a number of surgical procedures, but neurosurgical procedures are not among those tracked at this time. 12 Given our findings, we plan to modify our antibiotic prophylaxis protocol and eliminate PPSAs in other populations to improve our quality of care and prevent complications. Study Limitations Of course, our study has limitations. The number of cases of SSI, CDI, and resistant infections was small in both PPSAs and non-ppsas groups, making it difficult to draw firm conclusions regarding true superiority within the non- PPSAs group. Given the incidence of resistant infections in our study, we estimate that a noninferiority study would require 363 patients in each arm to detect a significant difference at an alpha of 0.05 with a power of To achieve this degree of certainty that there is a significant difference between rates of CDI in the 2 groups, we estimate that 549 patients would be needed in each arm. There were significantly more males in the PPSAs group than in the non-ppsas group (55% vs 40%), but the number of males and females who developed SSI, resistant infections, and CDI was similar, so there is no reason to believe that patient sex impacted our findings. Additionally, the PPSAs group had significantly more hemorrhages than the non-ppsas group (28% vs 6%), but among those who developed SSI, resistant infections, and CDI, the number of patients with hemorrhages was the same as the number with benign tumors. Thus, we do not believe that this factor affected our results. It is also important to note that we may have overlooked patients who had postoperative infections and were seen at another hospital; however, we expect that these patients would have ultimately followed up with the surgeon who performed their procedure, so this information would have been documented in our electronic medical record. Our cost analysis is based on the findings of prior studies and does not represent the exact cost for our patients. Addi- 911

5 A. Lewis et al. tionally, the antibiotic cost analysis is based on the assumption that patients were given only 1 g of vancomycin. This is not always the case, but our database did not include dosage information. Thus, we may have significantly underestimated the cost of antibiotics, as a premixed to 1500-mg bag of vancomycin costs $15 $19 at our institution. Conclusions After discontinuing PPSAs for patients with subdural or subgaleal drains at our institution, we did not observe an increase in the frequency of SSI. We did, however, note that there was a decrease in the frequency of CDI and the growth of resistant organisms. It appears that not only can patients in this population do without PPSAs, but complications can be avoided when antibiotic use is limited to 24 hours after surgery. The discontinuation of PPSAs proved to be a cost-reducing measure, which is a significant finding given the current climate of inflated health care spending. References 1. Abdelsattar ZM, Krapohl G, Alrahmani L, Banerjee M, Krell RW, Wong SL, et al: Postoperative burden of hospitalacquired Clostridium difficile infection. Infect Control Hosp Epidemiol 36:40 46, Abu Hamdeh S, Lytsy B, Ronne-Engström E: Surgical site infections in standard neurosurgery procedures- a study of incidence, impact and potential risk factors. Br J Neurosurg 28: , Akalin HE: Surgical prophylaxis: the evolution of guidelines in an era of cost containment. J Hosp Infect 50 Suppl A:S3 7, Akiyama Y, Miyazaki T, Daisu M, Yamamoto Y, Shinguu T, Maruyama N, et al: [Clinical efficacy and cost benefit of oral antimicrobial prophylaxis with levofloxacin on neurological surgery.] No Shinkei Geka 34: , 2006 (Jpn) 5. Alleyne CH Jr, Hassan M, Zabramski JM: The efficacy and cost of prophylactic and perioprocedural antibiotics in patients with external ventricular drains. Neurosurgery 47: , Bratzler DW, Dellinger EP, Olsen KM, Perl TM, Auwaerter PG, Bolon MK, et al: Clinical practice guidelines for antimicrobial prophylaxis in surgery. Am J Health Syst Pharm 70: , Centers for Disease Control: CDC/NHSN Surveillance Definitions for Specific Types of Infections. ( cdc.gov/nhsn/pdfs/pscmanual/17pscnosinfdef_current.pdf ) [Accessed April 15, 2016] 8. Dashti SR, Baharvahdat H, Spetzler RF, Sauvageau E, Chang SW, Stiefel MF, et al: Operative intracranial infection following craniotomy. Neurosurg Focus 24(6):E10, Dellit TH, Chan JD, Fulton C, Pergamit RF, McNamara EA, Kim LJ, et al: Reduction in Clostridium difficile infections among neurosurgical patients associated with discontinuation of antimicrobial prophylaxis for the duration of external ventricular drain placement. Infect Control Hosp Epidemiol 35: , Flibotte JJ, Lee KE, Koroshetz WJ, Rosand J, McDonald CT: Continuous antibiotic prophylaxis and cerebral spinal fluid infection in patients with intracranial pressure monitors. Neurocrit Care 1:61 68, Fridkin SK, Srinivasan A: Implementing a strategy for monitoring inpatient antimicrobial use among hospitals in the United States. Clin Infect Dis 58: , Joint Commission: Improving America s Hospitals: The Joint Commission s Annual Report on Quality and Safety ( TJC_Annual_Report_2013.pdf) [Accessed April 15, 2016] 13. Joundi RA, Wong BM, Leis JA: Antibiotics just-in-case in a patient with aspiration pneumonitis. JAMA Intern Med 175: , Korinek AM, Baugnon T, Golmard JL, van Effenterre R, Coriat P, Puybasset L: Risk factors for adult nosocomial meningitis after craniotomy: role of antibiotic prophylaxis. Neurosurgery 59: , Kourbeti IS, Vakis AF, Ziakas P, Karabetsos D, Potolidis E, Christou S, et al: Infections in patients undergoing craniotomy: risk factors associated with post-craniotomy meningitis. J Neurosurg 122: , Lemaire A, Dombrovskiy V, Batsides G, Scholz P, Solina A, Brownstone N, et al: The effect of Clostridium difficile infection on cardiac surgery outcomes. Surg Infect (Larchmt) 16:24 28, Little AS, White WL: Prophylactic antibiotic trends in transsphenoidal surgery for pituitary lesions. Pituitary 14:99 104, Magee G, Strauss ME, Thomas SM, Brown H, Baumer D, Broderick KC: Impact of Clostridium difficile-associated diarrhea on acute care length of stay, hospital costs, and readmission: A multicenter retrospective study of inpatients, Am J Infect Control 43: , May AK, Fleming SB, Carpenter RO, Diaz JJ, Guillamondegui OD, Deppen SA, et al: Influence of broad-spectrum antibiotic prophylaxis on intracranial pressure monitor infections and subsequent infectious complications in head-injured patients. Surg Infect (Larchmt) 7: , Murphy RKJ, Liu B, Srinath A, Reynolds MR, Liu J, Craighead MC, et al: No additional protection against ventriculitis with prolonged systemic antibiotic prophylaxis for patients treated with antibiotic-coated external ventricular drains. J Neurosurg 122: , Rebuck JA, Murry KR, Rhoney DH, Michael DB, Coplin WM: Infection related to intracranial pressure monitors in adults: analysis of risk factors and antibiotic prophylaxis. J Neurol Neurosurg Psychiatry 69: , Roberts RR, Hota B, Ahmad I, Scott RD II, Foster SD, Abbasi F, et al: Hospital and societal costs of antimicrobial-resistant infections in a Chicago teaching hospital: implications for antibiotic stewardship. Clin Infect Dis 49: , 2009 Disclosures The authors report no conflicts of interest concerning the materials or methods used in this study or the findings specified in this paper. Author Contributions Conception and design: Lewis, Pacione. Acquisition of data: all authors. Analysis and interpretation of data: Lewis. Drafting the article: Lewis. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Lewis. Statistical analysis: Lewis, Hill. Correspondence Ariane Lewis, Division of Neurocritical Care, Department of Neurology, NYU Langone Medical Center, 530 First Ave., HCC- 5A, New York, NY ariane.kansas.lewis@gmail. com. 912