Evaluation of bacterial contamination on surgical drapes following use of the Bair Hugger^® forced air warming system

Abstract

This pilot study determined the rate of bacterial contamination on surgical drapes of small animal patients warmed intra-operatively with the Bair Hugger^® forced air warming system compared to a control method. Surgical drapes of 100 patients undergoing clean surgical procedures were swabbed with aerobic culturettes at the beginning and end of surgery. Samples were cultured on Trypticase soy agar. Contamination of the surgical drapes was identified in 6/98 cases (6.1%). There was no significant difference in the number of contaminated surgical drapes between the Bair Hugger^® and control groups (P = 0.47).

Hypothermia is commonly encountered as a sequel to general anesthesia in veterinary small animal surgical patients (1). Maintenance of normothermia during anesthesia reduces rates of surgical site infection, reduces mortality, and decreases the length of hospital stay in human surgical patients (2,3). The Bair Hugger^® forced air warming device was developed in 1987 originally to warm patients during the postoperative period. It is now widely used in human and veterinary anesthesia to maintain normothermia during the perioperative period. In veterinary medicine, the Bair Hugger^® has been shown to be effective in maintaining patient temperature within 2°C to 4°C of normal during anesthesia (4). In a comparison of 4 intraoperative warming devices, forced air warming was shown to be effective at maintaining patient temperature, second only to a group warmed by an electric heating pad and warm water bottles (1). Although accepted to be an effective measure to maintain normothermia, controversy exists as to whether forced air warming systems increase the risk for surgical site infection (5–7).

To the authors’ knowledge, there are no studies in veterinary medicine that evaluate surgical site contamination from use of the Bair Hugger^® forced air warming system. The goal of this study was to compare the rates of bacterial contamination on the surgical drapes of small animal patients with and without the use of the Bair Hugger^® forced air warming system. The null hypothesis was that there would be no significant difference in the frequency of contaminated surgical drapes when the Bair Hugger^® system was used compared to the control method for intraoperative warming.

One hundred canine patients undergoing clean surgical procedures were enrolled into the study between September of 2011 and March of 2012. Patients were randomly assigned to the Bair Hugger^® (BH) group or the control group (CTL) using a coin toss. Patients were excluded if they weighed less than 5 kg, had evidence of pyoderma, or were receiving concurrent antibiotic therapy. Patient samples were excluded from analysis if the pre-operative samples were positive for bacteria.

The study group patients were warmed intraoperatively with the BH forced air system and a circulating warm water blanket placed beneath the patient. Each patient in the BH group received a new Bair Hugger^® perforated blanket at the beginning of anesthesia. One of 3 blanket models (55577, 53777, 53077, Arizant Health Care, Eden Prairie, Minnesota, USA) was used. The blanket model that covered the largest amount of non-clipped body surface area was chosen for each patient. Blankets were placed over each patient prior to final preparation of the surgical site. Two forced air units (model 505; Arizant Health Care), were designated to the operating suites and rotated among 3 operating rooms. A new filter was placed in each device prior to the beginning of the study. Patients in the control group were warmed intraoperatively by a circulating warm water blanket placed beneath the patient and with 3 warm water bottles heated to 41°C.

The patients were positioned in dorsal, lateral, sternal, or dorsolateral oblique positioning and draped routinely as required for the procedure being performed. An antimicrobial surgical incise drape (Ioban 2; 3M company, St. Paul Minnesota, USA) was used over the surgical site at the preference of the attending surgeon. After patient drapes were in place, dry, sterile rayon-tip swabs (BBL CultureSwab; Becton, Dickinson and Company, Sparks, Maryland, USA) were used to swab the surgical drapes. Swabbing was initiated from the margin of the paper or antimicrobial incise drape adjacent to the incision and continued laterally 6 cm and circumferentially around the surgical site. This sample was designated as the pre-operative sample (PRE) and was collected immediately prior to skin incision. For BH samples, the PRE sample was collected prior to turning the BH unit on. The surgical procedure was then performed routinely. A second sample swab was obtained as described for the PRE sample after placement of the last skin suture or staple. This sample was designated as the post-operative sample (PST). Immediately after collection, the sample swabs were cut mid-shaft with sterile mayo scissors and placed into sterile red top tubes containing 4 mL of sterile saline. Samples were then plated with a calibrated 1 μL loop (#3730, Science Center, Santa Fe, New Mexico, USA) onto combination plates of Trypticase soy agar with 5% sheep blood and MacConkey agar (#221289, Trypticase Soy Agar plates; Becton, Dickinson and Company). All samples were plated within 1 h of collection when possible. If a delay to plating was anticipated, then samples were refrigerated at 2°C to 4°C within 1 h of collection until plating was done. All samples were plated within 14 h of collection. Plates were incubated for 48 h at 35°C to 37°C. Bacterial growth was recorded as number of colonies formed. Bacterial identification to the genus level was performed on samples positive for bacterial growth.

Contamination of the surgical drape was defined as a positive bacterial culture from the PST culture swab.

The response variable in our study was bacterial contamination. The factors that could affect contamination were: method of intraoperative warming (BH or CTL), patient variables (age, weight, gender), patient positioning, type of procedure performed (orthopedic, soft tissue, or neurological), surgeon, number of surgeons (1 to 2 or 3+), number of total operating room personnel (1 to 5 or 6 to 9), operating room used, total anesthesia time, patient temperature at start and end of anesthesia (> 36.6°C or < 36.6°C), total surgical time, and the use of an antimicrobial incise drape on the surgical site. Univariate Chi-square testing was used to determine if the surgeon or operating room used had an effect on contamination. Any association of all other factors on contamination was quantified by means of multiple logistic regression. Multicollinearity was assessed by means of variance inflation factor (VIF); all VIF were < 2.5. Linearity was assessed by means of the Box Tidwell approach; age and body weight (kg) failed the linearity assumption and were transformed to binary variables. All variables were initially entered into the logistic equation and deleted according to highest P-value. Univariate analysis was also performed to determine if any factor was significantly associated with contamination. The Wilcoxon rank sum test was used to determine if there was a significant difference between bacterial counts in the BH and CTL population. Significance was set at P < 0.05.

One hundred canine patients were enrolled into the study. Samples from 2 patients were excluded from analysis due to positive bacterial culture on the PRE sample. Samples from 98 patients were used for statistical analysis.

Contamination of the surgical drapes occurred in 6/98 cases (6.1%). Contamination of the surgical drapes occurred 4 times in the BH group and 2 times in the CTL group (Table 1). There was no significant difference in the number of contaminated surgical drapes between the BH and CTL groups (P = 0.47). There was no significant difference in the bacterial colony counts between the BH and CTL groups (P = 0.35). Patient variables of age, weight, and gender had no significant effect on contamination (P = 0.13, 0.38 and 0.73, respectively). The variables of patient positioning, type of procedure performed, and number of scrubbed personnel had no effect on contamination (P = 0.94, 0.97, and 0.49, respectively). Additionally, total anesthesia time and the use of an antimicrobial incise drape also had no significant effect on contamination (P = 0.82, 0.30). Bacteria identified from contaminated samples included Staphylococcus from 5 samples and Micrococcus from 1 sample.

Table 1.

Contamination in relation to types of surgical procedures performed

Surgery type	Total number	Number Bair Hugger®	Number control	Contamination Bair Hugger®	Contamination control
Orthopedic	76	46	30	4	2
Soft tissue	13	4	9	0	0
Neurological	9	8	1	0	0

The overall incidence of contamination of the surgical drapes was 6.1%. There was no significant difference between the frequency of contamination or the bacterial colonies grown in the BH or CTL groups. Additionally, no patient or environmental factors were significantly associated with contamination of the surgical drapes. The most common bacterial contaminant was Staphylococcus which was found in 5 of 6 cases. These bacteria are likely to be normal flora from the patient skin and hair follicles or could represent contamination from human personnel. Micrococcus was noted in 1 case and likely originated as a contaminant from the patient’s skin or dirt residing on the patient’s hair coat.

Limitations of this study include the small number of samples that were positive for contamination, potentially creating a type II error. Additionally, the low incidence of positive samples in our study may due to the low sensitivity of bacterial detection, set at 4000 colony-forming units per sample swab. Further studies with larger sample sizes are warranted to further investigate rate of contamination with the BH forced air warming system. In addition, future studies should focus on less sample dilution and plating of a larger sample volume for a higher sensitivity in detecting contamination.

In conclusion, the overall incidence of bacterial contamination of the surgical drapes in our study was low (6.1%). We did not detect a significant difference in bacterial contamination of surgical drapes with use of the BH compared to the control. The clinical relationship between contamination of surgical drapes and surgical site infection (SSI) is unknown and beyond the scope of this study. Further studies with larger sample sizes are warranted to investigate the risk of surgical site contamination with forced air warming devices. CVJ