Risk Factors and Incidence for HAI in Elderly Patients Undergoing Neurosurgery in a Single Center

Background: Health care-associated-infection (HAI) is an important issue in neurosurgery department. With the aging population, we are receiving an increasing number of elderly patients who have a higher risks of HAI. AIM: We Methods: This was a case-control study. We took review of the medical records of patients aged 60 and older who 2016 single centre. Patients with infection and intubation before admission were excluded. Risk factors and outcomes were analysed, including age, sex, activities of daily living (ADL) score, smoking history, emergency admission status, ICU stay, ventilator use, ventilation time, diabetes, coronary heart disease, immune system disease, consciousness at admission, albumin decrease, blood sugar, unplanned second operation, indwelling venous catheter, external ventricular drainage (EVD), discharge ADL, length of stay (LOS), and medical costs. Results: A total of 1757 geriatric patients over 60 years underwent neurosurgical procedure were enrolled. Their age ranged from 60 to 93 years, with an average of 67.56 ± 6.16 years, including 1083 males(61.6%)and 674 females(38.4%).Among 1757 patients in 14703 hospital days, 95 cases of infection were diagnosed with HAIs in 78 patients, The HAI rate was 4.4% (78/1757) and HAI case rate was 5.4%(95/1757),incidence of HAI in neurosurgical geriatric patients was 6.5 ‰ . Of these 95 infections, the most common types were pneumonia (65.3%), followed by primary bloodstream infection (13.7%), central nervous system infection (8.4%), urinary tract infection (6.3%), and wound infection (6.3%). Multivariable logistic regression showed that the independent risk factors for HAI in elderly patients were ICU stay, mechanical ventilation, plasma albumin below 35g/L, unplanned reoperation, and venous catheterization. Conclusions: To prevent HAI, we should improve the ICU standards, decrease ventilation time, improve the nutrition supply, and dynamically evaluate the necessity of deep venous catheters and remove them as early as possible. Such measures could not only help reduce HAI and shorten hospital stays for elderly patients, but also decrease medical insurance cost for the government.


Introduction
Health care-associated-infection (HAI) is an important issue in hospital management. Previous studies have shown that cerebrospinal uid (CSF) leakage, external ventricular drainage (EVD), infratentorial surgery, diabetes, the nature of the surgery, incision type, operation timing, and operation time are risk factors for intracranial infection after neurosurgery [1]. In addition, age, a history of chronic diseases, hospital stay, and invasive procedures are also considered risk factors [2]. However, few studies have speci cally reported risk factors for elderly patients. The total number of older people (de ned as 60 years of age and over) worldwide is expected to increase from 605 million in 2000 to 1.2 billion by 2025 [3], and the morbidity of elderly infections is 18.57% [4]. HAI is an important issue that we focus on in the neurosurgical department, especially for elderly patients. This is important for the prevention and control of HAI after neurosurgery [5]. The aim of this study was to determine the rate and risk factors of HAI in elderly patients in the neurosurgery department, and the outcomes.

Subjects
Using the electronic database in hospital, consecutive patients who underwent neurosurgical operations between 1 January and 31 December 2016 at Xuanwu Hospital Capital Medical University were retrospectively identi ed. Hospital clinical data were extracted from 1757 patients. This study was approved by the Xuanwu Hospital Ethics Review Committee.

Risk factors and outcomes
Univariate variable analyses of the risk factors, including age, sex, ADL, smoking history, emergency admission, ICU stay, ventilator usage, ventilation time, history of diabetes, history of coronary heart disease, history of immune system disease, consciousness at admission, low serum albumin level, high blood sugar, need for an unplanned second operation, and venous catheters, were conducted. The patient's short-term outcome was assessed by ADL at discharge, length of stay in hospital (LOS), and hospital expenses.

Statistical analysis
Epi Data 3.0 was used for data entry and system logic error detection. Statistical analysis was performed using SPSS 22.0. The count data are presented as frequencies (percentages), and the differences between groups were assessed using the χ2 test. Normally distributed continuous data are presented as the mean ± standard deviation, and nonnormally distributed continuous data are presented as the median (P 25 , P 75 ). When the data were normally distributed and the between-group variance was equal, differences between groups were assessed using t-tests, while the rank sum test was used when the data showed a biased distribution. Univariate analysis was used to perform preliminary variable selection using an α level of 0.1. On the other hand, logistic stepwise regression analysis was used to analyse independent risk factors for HAI in elderly patients who underwent neurosurgery. Differences with P < 0.05 were considered statistically signi cant.

Patients
A total of 1757 geriatric patients over 60 years underwent at least one neurosurgical procedure were enrolled in a single center during this study period. Their ages ranged from 60 to 93 years old, average age was 67.56 ± 6.16. 1083 (61.6%) were male, and 674 (38.4%) were female.
Prevalence and type distribution of health care associated infections A total of 78 patients with 95 infection cases were detected among 1757 patients in 14703 hospital days. The HAI rate was 4.4% (78/1757) and HAI case rate was 5.4%(95/1757), incidence of HAI was 6.5‰(95/14703) in neurosurgical geriatric patients. Consequently, 0.96% (17/1757) patients had at least two site infections, while there were 1679 non-infected cases in this study (Table 1, Fig. 1). Pneumonia (65.3%), primary bloodstream infection (13.7%), and central nervous system infection (8.4%) were most common, followed by urinary tract infection (6.3%) and wound infection (6.3%). Univariate analysis of HAI Univariate analysis found that poor ADL score on admission, emergency admission, stay in the ICU, ventilation, mechanical ventilation over 48 hours, GCS less than 15, albumin below 35 g/L, high glucose, an unplanned second operation, venous catheterization and a long hospital stay were associated with HAI in neurosurgical patients over 60 years old (Table 2).

Outcomes
The median ADL at discharge in HAI group were 20, while Non-HAI group were 90. Outcomes of the HAI group were obviously worse than the non-HAI group including, discharge ADL, Length of stay (LOS), and total expense in hospital. (Table 4, Fig. 2). HAI group's length of stay in hospital is 3.08 times longer than non-HAI group. Hospital cost in HAI group is 178802RMB ,3.3 times more than non-HAI group, which is 54243RMB.

Infection rates
HAIs often occur in elderly patients after major surgery. Previous studies have shown that the postoperative infection rate in the neurosurgery department ranges from 3.52-17.97% [7,8]. According to the present study, in the group of patients over 60 years old, the postoperative infection rate was 4.4% (78/1757),as while as case rate was 5.4%, after neurosurgical procedures. Our study found that incidence rate of HAI in elderly patients after neurosurgery were 6.5‰. It means that in everyday, average 6.5‰ elderly patients got infectious after neurosurgery.

Infection types
According to studies in Europe, the most common infection types are pneumonia, surgical-site infections, and gastrointestinal infections [7]. Other studies showed that pneumonia, bloodstream infection and central venous infection were the most common types [9]. The ndings of this study were similar to the other studies for the types of infection, including pneumonia (65.3%), primary bloodstream infection (13.7%), central nervous system infections (8.4%). [2,[10][11][12].

Factors associated with perioperative period infection
A study focusing on elderly patients in hospital showed that HAI occurred in 3.52% of patients, of whom 80% were in ICU, and the infection site was mainly the lower respiratory tract (61.7%) [13,14]. The authors of that study concluded that urinary catheterization and intubation are independent in uencing factors. Another study found that the infection rate after neurosurgical procedures was 12.0%, which occurred an average of 7 days after the operation [10]. This phenomenon indicated that the rst week after operation is important for infection control. In this study, 0.91% of patients required at least one reoperation after surgery. The original disease also in uences the situation. It has been proven that patients with brain tumours, hydrocephalus, or cerebral vascular disease are more likely to become infected [15]. Other factors, including age, operation duration, reoperation, and requiring a blood transfusion have been shown to be relevant. Another report revealed that the postoperative infection rate of elderly brain tumour patients was 7.39%, which was associated with operative duration, administration of antibiotics, and chronic diseases [16]. In the present study, we found that there were no differences in age, sex, or a history of chronic disease between the infection group and the non-infection group. Other factors, including ADL, emergency admission, ICU stay, ventilation more than 48 hours, GCS ≤ 14, serum album 35g/L, high level of blood glucose, reoperation, and venous catheterization, were associated with infection in elderly patients who underwent neurosurgery.

SSI
Surgical site infections, including incision infections, meningitis, and brain abscesses, often occur 5-7 days after the operation [5]. SSI will negatively in uence the hospital days and expenses and even the prognosis of the patients. In this series, patients with HAI had an average of 18.5 days in hospital. EVD and a long operation duration are associated with SSI, and we noticed that some incision infections are directly caused by unfavourable suture techniques and postoperative management. Thus, the surgical team should be very careful regarding high-risk candidates and select suitable surgical methods to decrease the operation time and blood loss.

Possible impact of HAI
Worldwide, studies have shown that the average hospital stay of HAI patients is 30.4 days, compared with 10.6 days for non-infected patients. A meta-analysis showed that the morbidity of HAI was between 7% and 46%, the average hospital stay was increased by 5 to 21 days, and the average extra expenses reached up to 13000 USD [17]. Another study reported that emergency admission patients had more hospital stays and a higher mortality than normal patients [18]. In this study showed that HAI group's length of stay in hospital was 3.08 times longer than non-HAI group. The medium Hospital expense in HAI group was 178802RMB ,3.3 times more than non-HAI group, which was 54243RMB.

Limitations
This was a retrospective study in a single centre. Because the aim of this study was to determine the modi able factors associated with infection, we did not collect long-term follow-up data for all patients, so the long-term impact of HAI on patients and their prognosis is not yet clear and needs to be further studied.

Conclusions
We found a shorter ICU stay, shorter length of ventilator use, better nutrition supply for patients with albumin below 35 g/L, reducing unplanned reoperation, and careful evaluation of the necessity of venous catheters and removing them as early as possible could reduce the HAI rate in geriatric population. HAI prevention and control is a systemic task. We should recognize the high-risk factors and high-risk geriatric patients as early as possible. The prevention of hospital-acquired-infection can improve our clinical practice, enable patients to have better outcomes following neurosurgery and reduce the economic burden of medical insurance for the government. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Consent for publication
Not applicable