Post-Operative Shampoo Effects in Neurosurgical Patients: A Pilot Experimental Study

Study design-sample and sampling

A pilot experimental study was undertaken during December 2012.

Neurosurgical patients aged older than 18 years of age, admitted as scheduled cases undergoing a craniotomy, and who had given informed consent were enrolled in the study performed in a teaching hospital located in northern Italy.

Patients affected by conditions increasing the risk of postoperative SSIs [16] were excluded. Patients therefore were excluded who: Underwent a “clean-contaminated,” “contaminated,” or “dirty” neurosurgical procedure according to the Centers for Disease Control and Prevention criteria [4]; reported in their clinical history previous neurosurgical procedures, chemotherapy, radiotherapy treatments, recent infections, or judged at risk for infection (e.g., immunodeficiency) by clinicians; underwent spinal surgery or ventricular-peritoneal surgery, as well as those reporting ventricular drainage after surgery [17], or transferred to an ICU more than 48 h after the surgical procedure [2].

Hairless patients who had received total hair removal before the surgical procedure as well as those who were impaired cognitively were also excluded. Patient flow is represented in Figure 1. Eligible patients identified by a clinical nurse on a daily basis were assigned randomly to the exposed group, referred to as the hair-washing group (receiving a shampoo 72 h after the surgical procedure), or to the control group, referred to as the no hair-washing group (receiving standard dressing surveillance without shampooing), by a nurse researcher.

OPEN IN VIEWER

FIG. 1.

Patient flow.

Independent variable

The standards of care offered to both cohorts of patients were based on the literature available [18,19]. All patients received the same preparatory protocol for surgery. Patient hair was washed with an antiseptic iodine-containing shampoo the evening before surgery. Hair shaving, which was limited to the incision area (1 cm), was performed by clinical nurses once the location was defined just before the surgical procedure, with the help of the neuronavigator (StealthStation^®, Medtronic, Minneapolis, MN). In the operating room, the skin was prepared with povidone-iodine plus alcohol. At the end of the procedure, the surgical incision was sutured and disinfection was performed before the application of a sterile dressing.

All patients included were received cefazolin 2 g via infusion 30 min before the surgical incision and every 4 h during the procedure [20]. No included patient received post-operative antibiotics after the surgical procedure.

The dressing was changed after 72 h and on the fifth and seventh post-operative days. The surgical site was disinfected with a 10% povidone-iodine solution, which is effective against 90% of gram-positive and gram-negative bacteria, fungi, protozoa, yeasts, and some viruses within 90 sec of application [19].

Dressing change was performed using an aseptic technique and applying a ready-made dressing. The removal of sutures or clips was performed between the seventh and tenth day after surgery. In addition, the hair-washing group received a gentle shampoo 72 h after the surgical procedure, at the first dressing change, with a basic solution (pH 5.5) and subsequent hair drying with a clean towel or hair dryer available in the unit and stored appropriately in a clean manner.

Study end points

Three study end points were assessed in both groups: The degree of comfort as perceived by patients: A numerical rating scale (NRS) ranging from 0 (none) to 10 (maximum comfort) was administered after 72 h to both groups, after the dressing change for the no hair-washing group, and after the shampooing plus dressing change for the hair-washing group.

Second, Contamination of the surgical site, measured as the number of the colony-forming units (CFU) [21,22]. The effect of the post-operative shampoo on the increase of the number of CFU present in the surgical site was evaluated. Few studies have examined the role of the patient's endogenous flora in SSI and have suggested that there may be a correlation between bacterial counts on the skin and SSIs. Some authors have argued that the measurement of the contamination of the surgical site is considered an acceptable surrogate outcome of infection [23,24]. Moreover, Cronquist et al. [25] have demonstrated no association between pre-operative bacterial skin counts and SSI in 609 neurosurgical patients exposed to craniotomy and counted for their skin CFU flora at the site and subsequent SSI. However, in the case of post-operative shampooing, exogenous sources of contamination including personnel, the environment in which shampooing is performed (hospital bathroom, shared with several patients), and the materials used (hair dryer available in the unit) [15], may increase the risk of surgical site contamination.

Therefore, two samples were taken by rubbing the swab around and on the incision on the subcutaneous cell tissue of the wound edges. An Eswab system (Copan Diagnostics, Murrieta, CA) was used for the collection and transport of the specimens [26]. All specimens were collected by nylon flocked swabs, and each swab immediately eluted in a tube with a liquid transport medium (1 mL of modified liquid Amies), which maintained the viability of aerobic and anaerobic bacteria for up to 48 h at room and refrigerator temperatures. The colony count was therefore performed on a semi-quantitative culture. According to the available literature [22], a presence of more than 15 CFU was considered as contamination.

In the exposed group a wound swab was taken at T0 (after 72 h, when the dressing was removed), and at T1 (after the patients had received a shampoo, had their hair dried, and had the surgical incision disinfected with povidone-iodine solution just before applying the new dressing). In the control group, a wound swab was also taken at T0 (after removing the dressing but before the skin was disinfected) and at T1 (after disinfecting the skin and before applying the new dressing);

Third, the occurrence of SSIs within 30 d after the surgical procedure. According to the criteria of the U.S. Centers for Disease Control and Prevention a SSI was diagnosed with clinical data in the presence of one of the following cases: (1) A purulent discharge from the incision, (2) a serous discharge with a positive bacterial culture, (3) a deep or superficial incision abscess with or without a positive bacterial culture, (4) an incision swelling and erythema with or without pyrexia, meningitis/ventriculitis with positive bacterial culture or microorganisms observed on gram staining, all occurring within 30 d of the surgical procedure [10]. The data were collected by researchers who interviewed patients and collected data from the clinical records. Demographics (age, gender) as well as clinical data (co-morbidities, pharmacologic treatment), surgical procedure received (general or local anesthesia) and its duration (min), which may influence the occurrence of SSIs [22–25], were also collected from the clinical records.

Study approval

The study was approved by the internal review board of the teaching hospital. Patients gave their written informed content after having received appropriate information on the study aims. Data were collected and managed ensuring confidentiality.

Data analysis

Descriptive statistics were calculated with SPSS version 20 (IBM Inc., Armonk, NY), including frequencies, percentages and averages, standard deviations (SD), and 95% confidence intervals (CI). The analysis of the data was carried out in a blinded fashion. Differences, if any, between the groups, were calculated using the χ² test (Fisher exact test when appropriate) for categorical data, and the Mann-Whitney U test for continuous data according to the skewed nature of the variables under study. In the bivariate analysis, correlations were analyzed using the Pearson test (r). The level of statistical significance was fixed at p<0.05.

Participants

A total of 53 patients were included: 37 (69.8%) males and 16 (30.2%) females. The average age was 52 years (95% CI 47.4–56.5 years). Forty-one (77.3%) patients received general anesthesia and 12 (22.6%) received an awake craniotomy. Forty-one (77.3%) patients underwent surgery for a brain neoplasm and the remaining 12 (22.6%) for chronic subdural hematoma where a small craniotomy was performed. The length of the entire surgical procedure was on average 229.4 min (95% CI 188.15–270.12 min).

Twenty-five patients (47.2%) received a shampoo after 72 h and 28 (52.8%) received standard care. Hair drying was performed with a clean towel (7; 28%) or hair dryer (18; 72%) by the same clinical nurse according to the patient's preference. Table 1 presents the principal characteristics of the exposed and control group.

Comfort perceived by patients

Overall, patients reported an average score of 7.70 for comfort, ranging from 0 (none) to 10 (maximum) (95% CI 6.96–8.40). Patients who received a shampoo reported a similar level of comfort (average=8.04; SD 1.05) than those receiving standard care (average=7.3; SD 3.2), which was not statistically different (p=0.345).

The level of comfort was not correlated with age (r=−0.163; p=0.244); although there were significant differences in level of comfort between males (7.19; SD 2.89) and females (8.15; SD 2.23; p=0.042), between those who underwent general anaesthesia (7.95; SD 2.30) and awake craniotomy (6.30; SD 3.60; p=0.032), and between those undergoing surgery for removal of a brain tumor (8.10; SD 2.30) or treatment of a chronic hematoma (6.30; SD 3.10; p=0.04). No other significant differences emerged.

Surgical site contamination

Seven surgical sites of 53 (13.2%), six at T0 and only two at T1 among those patients received shampooing (T1), reported a culture of more than 15 CFU, therefore, these sites were considered contaminated. The only site contaminated after shampooing was treated with a hair dryer available at one unit. In terms of the occurrence of contamination, there were no differences between groups (Table 2).

Surgical site infection

All patients completed their 30-d follow-up. No infection occurred in either the shampooed or control group.

Study end points

Overall, our patients reported comparable levels of comfort, not significantly different related probably to the small sample size. These preliminary findings may be associated with the reduction of proritis caused by the presence of residues of blood, disinfectants, and adhesive dressings [28,29].

Females reported a significantly greater level of comfort, perhaps because of their hair length, which is usually longer; brain neoplasm patients also reported a significantly greater level of comfort, compared with those who underwent surgery for hematoma, possibly because the latter were mainly older, whereas other factors such as comorbidities and dependency in daily living activities may affect the overall level of comfort. Last, the significantly lower levels of comfort perceived by those patients who underwent awake craniotomy may be associated with the more demanding role played by patients during the surgical procedure, which may increase discomfort and fatigue [30].

Moreover, comfort was measured after 72 h; a subsequent measurement is recommended (e.g., after 96 h or later) in further studies, given that the consequences of non-hair-washing can become progressively more uncomfortable for patients over time, as both organic material and dirt increase.

There were no differences in surgical site contamination between groups. Immediately after removal of the dressing, the number of sites with more than 15 CFUs was similar, whereas after the shampoo and disinfection, only one site was contaminated in the exposed group and no sites were contaminated in the control group receiving standard care. The only site contaminated after having received shampooing was treated with the hair dryer available at the unit level, therefore, its use should be considered with care in the future, given that this may be a possible source of contamination.

As expected, given the low incidence of SSIs reported in the literature [2,4–6], our patients developed no infection by 30 d. A comparison of patients involved in our study with the only comparable study reported by Ireland et al. [14] revealed homogeneity in age, gender, surgical procedure, and clinical profile. Unlike Ireland et al. [14], all patients were followed for 30 d and none withdrew from the study. The findings for the main end point (SSIs) were homogeneous, whereas in terms of the comfort perceived by patients as a proxy indicator of pain our findings did not confirm those reported by Ireland et al. [14], who found that the exposed group reported significantly greater pain-related limitations affecting daily living activities.

Study limitations

The absence of previous studies in the field that would offer information concerning the sample size prompted this pilot study, which nevertheless was affected by several limitations that need to be addressed with larger multi-center experimental studies. A limited number of patients were included in the study; they showed homogeneity for demographic and clinical characteristics influencing the risk of developing SSIs. Therefore, these findings are applicable only to a specific homogeneous population and they should be generalized prudently for low-risk infection patients who undergo a clean surgical procedure, whereas a more diverse group of patients would be required in order to develop more generalizable evidence.

Surgical site contamination was considered an end point. Although, this measure is well established in other fields with good sensitivity and reliability with regard to SSIs [22], it has been not associated with an increased occurrence of SSIs after neurosurgical operations [25]. Contamination of the surgical site, was considered suggest an end point for two different reasons: (1) The shampooing procedure adopted in the hospital environment may increase the exogenous contamination of the surgical sites; and (2) neurosurgical SSI occurrence should be influenced by exogenous contamination. However, according to the literature, approximately 21,000 cases would be needed to have a 90% chance of identifying differences in occurrence at the 5% confidence level in a one-tailed test [6].

On the basis of the findings of this study, according to the difference in the number of contaminated sites between the exposed and the control group (4%), the sample size suggested further studies, ensuring a power of 80% (with a confidence interval of 95%), would be 416 patients in the exposed group and 416 in the control group.