Clinical features and risk factors of surgical site infections in HIV-negative patients with cryptococcal meningitis underwent ventriculoperitoneal shunt operations: a retrospective study

Background To investigate the clinical features and risk factors of ventriculoperitoneal shunt (VPS) associated surgical site infections (SSIs) in HIV-negative patients with cryptococcal meningitis (CM). Methods We retrospectively reviewed the medical records of HIV-negative patients with CM underwent VPS operation admitted to The Third Affiliated Hospital of Sun Yat-sen University in Southwest China over the past 7 years. Results 193 patients were included, of whom 25 (12.95%) had SSIs in 6 (median duration, 1–48 days) days after operation. Compared with patients without SSIs, patient with SSIs tended to be shorter preoperative stay. 52% patients in SSIs group and 25% patients in no-SSIs group underwent VPS operations within 3 days after admission (p = 0.017). Although body temperature and infectious indicators slightly elevated postoperative in both groups. The patients with SSIs experienced more fever; more central nervous system symptoms; higher PCT value and lower cerebrospinal fluid (CSF) glucose in contrast to the no-SSIs group. Multivariate regression analysis found a 2.653 fold increase in the risk of infection for every 1 °C increase in postoperative body temperature. Among the 25 patients, 9 patients had positive culture results, three samples reported to be oxacillin resistant coagulase-negative Staphylococci. Conclusions SSIs was one of the serious surgical complications after VPS operation. High body temperature, the occurrence of dizziness and headache, low postoperative hemoglobin are risk factors. Postoperative patients with high fever, high PCT and low CSF glucose should be paid more attention to.

Recently, CM has been increasingly observed in HIVnegative patients [6].
The main clinical manifestations of CM were fever and intracranial hypertension (ICH), including headache, nausea, and vomiting. Persistent ICH could lead to mental abnormalities, consciousness impairment, hearing loss, and visual loss [7]. Cerebral hernia may occur in severe conditions and leading to death, which is one of the major reasons for death in CM patients [8].
Ventriculoperitoneal shunt (VPS), as a solution of high intracranial pressure, is now also widely used in patients with CM-associated ICH [9]. Several studies conducted to verify the therapeutic effects of VPS in non-HIV CM patients confirmed that the placement of a VPS is helpful in decreasing ICH and fungal overload in non-HIV CM patients [9,10]. However, one of the serious surgical complications, surgical site infections (SSIs) which has a range of 1-39% incidence [11,12], would lead to re-implantation and longer duration of hospitalization, increased of hospitalization expense and even death [13][14][15]. Much efforts have been made to identify the characteristics and risk factors for SSIs of VPS operation, in an aim to help clinicians recognize the occurrence of infection earlier thus improve outcome. But most of these studies included patients with intracranial hypertension secondary to a variety of health conditions [16][17][18]. To the best of our knowledge, there are rarely studies describing the clinical manifestations, characteristics, and risk factors of VPS associated SSIs in HIV-negative CM patients. The manifestations of postoperative infection are nonspecific, including fever, neck resistance, cerebrospinal fluid (CSF) leukocyte change [18], which could be confused with the central infection symptoms of CM.
Thus, we performed a retrospective analysis of the episodes of VPS associated SSIs in patients with CM among HIV-negative population to describe the clinical characteristics and risk factors for VPS associated SSIs in patients with CM.

Study population
This is a retrospective case control study including all patients diagnosed with CM and had undergone VPS surgery from January 1st 2014 through 31th December 2020 in The Third Affiliated Hospital of Sun Yat-sen University. Excluded from the study were (1) pregnant/ lactating women; (2) death due to non-infectious factors within 7 days after operation; (3) other operations were performed at the same time with VPS; (4) HIV positive patients.
All patients were given antifungal medicine and other symptomatic therapy. The protocol was approved by the Ethics Committee of the Third Affiliated Hospital of Sun Yat-sen University.

Data collection
Hospital charts were reviewed with a standardized case-report form to retrieve demographic, clinical, radiographic, and laboratory data. Dual data entry by different operators was performed, with preprogrammed consistency checks. The following data were collected: demographics (sex, gender, age and history of diabetes); preprocedure/surgery time; the duration and type of antimicrobial prophylaxis use; body temperature, clinical manifestation, blood test [white blood cells (WBC), hemoglobin, platelets, blood albumin (ALB), C-reactive protein (CRP), and procalcitonin (PCT) levels], cerebrospinal fluid (CSF) profiles (CSF white blood cell count, CSF glucose, CSF protein, and cerebrospinal fluid pressure), preoperative and postoperative.

Definitions
Diagnostic criteria for CM were positive culture with cryptococcus in CSF, or positive for India ink staining of CSF smear [19] or known history of cryptococcosis, with clinical manifestations of intracranial hypertension.
An infection was considered to be related to the VPS surgery if at least 1 of the following 2 criteria was fulfilled (according to modified criteria for nosocomial infections of the Centers for Disease Control and Prevention [CDC]) [20]: (1) culture positive in the CSF or in wound secretion and the pathogen was interpreted as relevant, or (2) fever (temperature > 38.5 °C), positive culture of blood, symptoms that newly emerged or worse of ICH (headache or dizziness, neck stiffness, cranial nerve signs, or irritability) with no other explainable cause; physician initiation of an appropriate antimicrobial therapy for SSIs. The onset of infection was defined by the first positive culture, the initiation of an appropriate anti-infective treatment for the SSIs, or surgery at the site of the shunt (whichever occurred first).

Statistical analysis
SPSS statistics version 25.0 (IBM, Armonk, New York, USA) was used to analyze the data. Continuous variables were presented as Median ± Quartile Range (QR) for non-normal data, or Mean ± standard deviation (SD) for normal data, and categorical variables were summarized as frequencies. T test was used to compare normally-distributed, while Mann-Whitney U test or Wilcoxon signed rank test were used to compare non-normally distributed data, respectively. Categorical variables were compared using the McNemar test, Chi-squared test or Fisher's exact test. P values for which smaller than 0.05 were considered statistically significant. The multivariate logistic regression was used to analyze the risk factors for infection after VPS.

Demographic characteristics
A total of 207 CM patients performed on VPS surgery from January 2014 to December 2020 were identified (Fig. 1). Of these, 3 patients died or discharged for non-infection complication within 7 days after the surgery, 5 patients underwent other surgery at the same time, 4 patients diagnosed infection preoperative, and 2 patients lack of data. Therefore, 193 VPS surgery cases were included. According to the infection identification, 25 patients develop surgery-associated infection after operation, the incidence of SSIs associated with VPS operation is approximately 13%. The demographic characteristics of this population were shown in Table 1. 72% of the patients were male. The mean patient age was 45 years (range, 5-72 years). The mean duration of hospital stay was 40 days (range, 4-122 days). 52 patients were admitted to intensive care unit after surgery.

Comparison of clinical symptoms and laboratory indices before and after VPS
The results on preoperative and postoperative changes in clinical symptoms, body temperature, and laboratory characteristics were summarized in Table 2. Body temperature and clinical symptoms were estimated 1 day before and 1 day after the surgery. Intracranial pressure  and laboratory characteristics were compared with the last preoperative measurement to the first postoperative measure. After surgery, the body temperature slightly increased, and the percentage of ICH symptoms was slightly decreased while the intracranial pressure decreased approximately 50%. The inflammatory indices, including white blood cell (WBC) of blood, CRP, PCT, elevated to various degrees after surgery. In contrast, there was no significant difference in WBC counts of CSF postoperative. The CSF protein level visibly increased from 0.67 to 1.72 g/L.

Comparison of clinical symptoms and laboratory indices preoperative between infection group and non-infection group
Demographic data, surgery-related information and results of the blood and CSF examination preoperative in infection group and non-infection group were shown in Table 3. There was no significant difference in gender, age and complications between the infection group and noninfection group. Most symptoms, clinical signs and laboratory test results did not differ significantly between the two groups before surgery, except for the preoperative stay. More patients underwent operation within 3 days after admission in the infection group. According to the guiding principles of clinical application of antibiotics  in China, cefazolin or cefuroxime is recommended to be prophylactic antibiotics for VPS, and the duration of prophylactic antibiotic should not exceed 24 h. The antibiotic prophylaxis strategy were basically the same in the two group. Standard prophylactic antimicrobial agents (cefuroxime or cefazolin) were used in 56% patient in the infection group and non-infection group, meanwhile 60% patient were given the short duration of prophylactic antibiotics (up to 24 h).  Table 4. Body temperature would be 0.5 °C higher in the infection group. Intracranial pressure was slightly higher in the infection group, but without statistical significance. There was no significant difference in postoperative blood WBC and CRP between infection group and non-infection group, but the PCT of infection group was slightly higher than that of noninfection group. In the first postoperative cerebrospinal fluid examination, the number of leukocytes in the two groups was equivalent, while the CSF glucose in the infection group was significantly lower than that in the non-infection group. The CSF protein also increased slightly, but with no statistical significance. In the multivariate logistic regression analyses, there were finally two parameters significantly correlated with SSIs (

Discussion
VPS, which has been proposed to reduce the development of cerebral hernia, may improve the prognosis of CM patients with intolerable ICH. Most of the patients in this study achieving symptom remission significantly after surgery, the intracranial pressure decreased, the ICH symptoms such as dizziness, headache, nausea and vomiting were remarkably responsive to VPS which in common with the previous study [21], suggested that VPS surgery would be efficacy to the CM patients complicated with ICH. However, infection would be one of the major complications of VPS surgery. CSF shunt infections could present with few or no symptoms [18,22]. Especially in CM patient, the central symptoms of CM which was an infectious disease itself, may interfere with the diagnosis and treatment of CSF shunt infection. Most published studies of CSF shunt-associated infection mainly involved patients with various non-infectious diseases that cause ICH [12,17,18,23]. This study is the first report of SSIs associated with VPS operation in HIVnegative patients with CM.The rate of internalized CSF shunt infection in most studies has ranged from approximately 5% to 15% [17,18,22]. Similar to the findings, the infection rate of VPS in HIV-negative patient with CM in our study was 12.95%. Measures for prevention of SSIs associated with VPS operation include meticulous adherence to surgical and sterile technique (including topical antiseptic), short procedure times, and perioperative antibiotic prophylaxis [24,25]. Antibiotic-impregnated catheters may be beneficial, too. Use of systemic prophylactic antibiotics decreases rates of CSF shunt infection [26]. According to the guidelines, cefazolin or cefuroxime given 60 min prior to the incision is recommended for patients undergoing clean neurosurgical procedures, including VPS. Cefazolin should be redosed at 4-h intervals until the surgery is over. The IDSA guidelines for the management of healthcare-associated meningitis and ventriculitis suggest that antibiotics be continued for as long as 24 h postoperatively as this is the duration that has been studied [22]. In our study, all the catheters were vancomycin impregnated, and all the patients were given prophylactic antibiotics. But many surgeons hesitated to adhere to the choice of recommended antibiotics agent and the short-course prophylactic antibiotics. Our results demonstrated that application of advanced antibiotics such as ceftriaxone, or prolong the duration, would not reduce infection risk.
Compared with shunt-associated infection in the general population that usually manifests as nonspecific clinical signs and symptoms [18], the clinical manifestations of SSIs associated with VPS in patients with CM seemed to be more obvious, possibly due to the coexistence of CM or unknown immune system disorder in this population. Fever > 38 °C, headache and ICH symptoms were present in over 70% of patients. Although the surgery procedure would lead to a slight increase in body temperature, WBC of blood, CRP, PCT. However, compared with the non-infection group, the postoperative body temperature and PCT of the infection group were much higher, while the CSF-glucose was significantly lower. It suggested that we should consider the possibility of infection when these conditions occur simultaneously after VPS operation in CM patients.
In addition, the majority of VPS associated SSIs occurred within 1 week after surgery (over 60% of the patients). The mean time of infection development after surgery was 8.88 days (range, 1-48 days), which was shorter when compared with the results of other studies that conducted in another population [17,23]. A possible explanation might be that HIV-negative patient with cryptococcus meningitis may potentially have certain immunodeficiencies that are still undetectable, which therefore makes them more prone to infection and allows the infections manifest more quickly. This study had limitations due to its retrospective design. Firstly, the diagnosis of infection relied heavily on medical records. If these medical records were not seriously recorded, some infections might be missed. Secondly, missing data was inevitable due to its retrospective design. Thirdly, this study recorded patients' symptoms and laboratory test results during hospitalization, which partially reflected the short term outcome of HIV-negative patients with CM undergoing VPS operations. Other outcome evaluation methods liked Glasgow outcome score and long-term follow up were needed to accurately access the efficacy of VPS in this population. Nevertheless, we believe that our study is valuable and has several strengths. Firstly, this analysis covered a long period and included a relatively large number of patients. Secondly, the inclusion and exclusion of patients was carried out in strict accordance with the criteria.

Conclusion
The incidence of VPS associated SSIs in patients with CM was approximately 13%. High body temperature and low postoperative hemoglobin were risk factors for VPS associated SSIs. High postoperative body temperature, low postoperative hemoglobin, high PCT and decreased CSF-glucose might be the indictors for infection. Multivariate logistic regression analyses indicated high temperature postoperative (OR, 2.653; 95% CI 1.108-6.353), low postoperative hemoglobin (OR, 0.947; 95% CI 0.908-0.988) were risk factors for VPS associated SSIs.