The Effect of Preoperative Subcutaneous Fat Thickness on Surgical Site Infection Risk in Patients Undergoing Colorectal Surgery: Results of a Multisite, Prospective Cohort Study

Surgical site infection (SSI) is one of the most frequent postoperative complications among patients undergoing elective colorectal surgery. A multisite, prospective cohort study was conducted to investigate whether the thickness of subcutaneous fat (TSF) influences the occurrence of SSI in patients undergoing colorectal surgery. Participants included patients scheduled to receive colorectal laparotomy for colorectal cancer and who were under the care of a wound ostomy continence nurse at 17 participating general hospitals in Japan.Patients were not eligible to participate if they had undergone emergency surgery, reoperation, or laparoscopic surgery. Demographic, wound, and surgical data and American Society of Anesthesiologists (ASA) scale scores were collected and assessed, along with nutritional status, TSF, body mass index, and risk factors for SSI (ie, length of surgery and wound classification). The incidence of SSI and nutritional conditions was assessed weekly for 30 days after surgery. Of the 155 participants (mean age 68.9 ± 10.8 years, 53 [34.2%] of whom were women), 90 (58.1%) underwent rectal surgery, and the remaining 65 underwent colon surgery. Seventy-two (72, 46.5%) of the 155 patients underwent colostomy surgery; 24 (15.5%) developed a SSI. The mean onset of SSI was 7.3 ± 2.9 days after surgery and commonly observed in the stoma group when the ASA score was 3 or higher (P = 0.02). Patients who developed SSI resumed oral dietary intake later than those without SSI (7.4 days versus 4.6 days, P = 0.02). Multivariate analysis indicated TSF >15 mm (P = 0.01), Alb level <3.5 g/dL at postoperative days 14 (P = 0.03) and 21 (P = 0.02), and total protein level <6.8 g/dL at postoperative day 7 (P = 0.02) were statistically significantly correlated with SSI occurrence. These results suggest preoperative TSF and preoperative and postoperative serum albumin levels are independent risk factors for SSI in patients undergoing colorectal surgery, confirming current recommendations to maximize the preoperative nutritional status of patients in order to prevent major complications. The results also confirm obesity/TSF should be assessed preoperatively to ascertain the risk of SSI in patients undergoing colorectal surgery.  

Surgical site infection (SSI) is the most common health care-related infection, affecting approximately 31% of all hospitalized patients regardless of cause of admission.¹ According to the Japanese Healthcare Associated Infections Surveillance² (JHAIS) of 103 institutions and 6818 cases, between 1998 and 2011 the SSI rates among patients undergoing colorectal, colon, and rectal surgeries were 16.33%, 15.03%, and 19.02%, respectively. Higher rates of SSI associated with colon and rectal surgeries have been reported in Japan than in the United States according to the data provided by the National Healthcare Safety Network (NHSN) Report.³ However, the data collection period differed between the 2 countries in that study; patients were tested for SSI for an average of 5.7 days after surgery⁴ in the United States and for 29.8 days after surgery⁵ in Japan. This methodological difference may have increased identification of SSI in the Japanese study.

Among all surgical procedures, several risk factors of SSI have been identified, including length of surgery >180 minutes, wound classification according to the Centers for Disease Control and Prevention (CDC) guideline higher than class III, and an American Society of Anesthesiologists (ASA) score >3.^6-8 Because these risk factors are recognized, efforts have been made to control them nationally and internationally. Furthermore, several cohort studies^9-15 have indicated stoma placement and obesity may be associated with SSI risk. Utsumi et al¹¹ prospectively examined 12,015 Japanese patients who underwent gastrointestinal surgery and found stoma placement was a risk factor for the development of SSI. A prospective, observational multicenter study¹² (N = 61) conducted in Spain supports Utsumi’s findings; stoma placement was found to be an independent risk factor for SSI. Tang et al¹⁵ prospectively examined 2809 patients who underwent elective colorectal surgery and stoma placement and found an ASA score >2 was a risk factor for SSI. A prospective cohort study¹³ of 556 patients who underwent colorectal surgery identified stoma placement as an independent risk factor for SSI.

Body mass index (BMI) is a well-known indicator of obesity. However, because BMI is calculated from the patient’s height and weight, it may not accurately reflect thickness of subcutaneous fat tissue.

Recent SSI studies have focused on thickness of subcutaneous fat (TSF), which can be more objectively measured than BMI by using computer tomography.^16,17 A cohort study¹⁶ (N = 655) of patients who underwent midline laparostomy reported increased subcutaneous fat (odds ratio [OR] = 1.76 per 10% increase) was an independent risk factor for superficial incisional SSI. A cohort study¹⁷ examined 152 patients with colorectal surgery and found TSF was an independent risk factor for SSI.

However, a comprehensive investigation of the association between SSI and TSF has not been conducted. The purpose of this prospective cohort study was to determine whether perioperative TSF in patients undergoing selective colorectal laparotomy is associated with SSI.

Study settings. The institutional level inclusion criteria used in the prospective comparison in this cohort study stipulated participating institutions should be hospitals 1) where certified wound, ostomy, and continence (WOC) nurses care for digestive organ cancer patients; 2) where infection control practitioners manage infection across the institution; and 3) that participate in the Japan Nosocomial Infections Surveillance¹⁸ (JANIS) or JHAIS² studies conducted in accordance CDC guidelines⁸ for SSI prevention. WOC nurses working at 25 institutions that met the inclusion criteria were mailed invitation letters; 8 institutions declined participation. WOC nurses at the 17 remaining hospitals recruited consecutive patient participants who were scheduled to undergo a colorectal laparotomy for colorectal cancer. Patients were excluded from recruitment if they had undergone emergency surgery, reoperation, or laparoscopic surgery. All surgeries performed in the participating institutions were managed according to the World Health Organization (WHO) Guidelines for Safe Surgery¹⁹ and the CDC guidelines for SSI prevention and management.⁸ Thus, the perisurgical procedures in each hospital were considered standardized with regard to environmental, perioperative, and intraoperative procedures.

Ethical considerations. The study procedures were reviewed and approved by the internal review boards of the Graduate School of the International University of Health and Welfare and the 17 participating hospitals. Oral and written informed consent was obtained from each participant.

Measurements. Nutritional status, TSF, and BMI were assessed, along with SSI risk factors that included duration of the surgery, wound classification according to CDC guidelines, and ASA score >3. To evaluate nutritional status, subjective global assessment (SGA) was used, which comprised 3 categorical states: well nourished, moderate or suspected malnutrition, and severe malnutrition.²⁰ The SGA score was determined by nutritionists at admission.

TSF was measured using CT scan data initially obtained to confirm the cancer diagnosis before surgery. Subcutaneous tissue thickness under the umbilicus was automatically calculated by WOC nurses using the measuring tool of the CT machine. A likelihood-ratio test determined the TSF cutoff point as 15 mm.

BMI was categorized in accordance with the WHO guidelines²¹: underweight <18.5 kg/m²; normal, between 18.5 and 25 kg/m²; and overweight >30 kg/m².

Risk factors. Surgical duration was considered a risk factor if it exceeded the 75th percentile of the NHSN or JANIS standard. Wound classification was considered a risk factor if the wound was assessed as either contaminated or purulent/infected. The SSI risk score has been widely utilized by the NHSN and in other countries.^3,7,8

Data collection. Demographic data collected included age, gender, smoking status, history of diabetes, site and duration of surgery, surgical wound classification, stoma replacement, and the date oral intake resumed. Baseline data (including TSF, BMI, and serum Alb [albumin] and total protein [TP] levels) were collected before admission. Blood laboratory variables were followed-up weekly until 30 days after surgery. BMI was reassessed at postoperative days 7 and 14. SSI and the degree of contamination were diagnosed by the board-certified Infection Control Pharmacy specialist according to the NHSN criteria³ by physicians. All other variables were collected by WOC nurses. Once the participants were discharged, they were followed-up for 30 days after surgery. Data were collected from August 2011 to May 2013 and entered directly into a computer database. 

Statistical analysis. Normality of variables was verified using the Shapiro-Wilk test. The chi-square test was used to determine demographic differences between the patients who developed SSI and those who did not. Analysis of variance (ANOVA) was used to determine the correlation among the 3 SGA states to Alb level, TP level, TSF, and BMI. The Mann-Whitney test was used to compare the Alb level, TP level, TSF, and BMI. Significant variables derived from the t-test were included as independent variables in the univariate analysis. The final analysis was a multivariate analysis of variables found significant in the univariate analysis. In addition, 95% confidence intervals (CIs) also were calculated. Power analysis was performed before data collection with power and sample size calculations. The required minimum sample cases were 104 (power = 0.8, alpha = 0.1). All data were analyzed using standard statistical software (SPSS 21.0 for Windows, Chicago, IL).

Participant demographics. In this study, 155 patients (53 [34.2%] women) were enrolled and followed-up until 30 days after surgery (see Table 1). Mean age (68.9 ± 10.8 years) did not differ significantly between the 2 groups. Of the 155 patients, 34.2% were women; 90 (58.1%) underwent rectal surgery, and the remaining 65 underwent colon surgery. Seventy-two (72, 46.5%) of the 155 patients underwent colostomy surgery. 

SSI rates. Nine (9) patients (5.8%) who underwent colon surgery and 15 patients (9.7%) who underwent rectal surgery developed SSI for an overall SSI rate of 15.5%. Among the 24 patients with SSI, 16 underwent colostomy surgery (22.2%; chi 0.67; P = 0.04). The mean onset of SSI was 7.3 ± 2.9 days after surgery. SSI was commonly observed in the stoma group when the ASA score was 3 or higher (P = 0.02).

Factors associated with SSI. The SGA score was lower in patients who underwent colostomy surgery than in those who did not (P <0.01); however, in colostomy surgery patients only, no significant relationship was found between SGA score and SSI incidence (P = 0.28). 

TSF >15 mm (P = 0.01), Alb level <3.5 g/dL at postoperative days 14 (P = 0.03) and 21 (P = 0.02), and TP level <6.8 g/dL at postoperative day 7 (P = 0.02) were statistically significantly correlated with SSI occurrence (see Table 2). The patients who developed SSI resumed oral dietary intake later than those without SSI (7.4 days versus 4.6 days, P = 0.02). The univariate analyses of stoma construction and low TSF, Alb level, and TP level as independent variables confirmed these variables were significantly associated with SSI. Logistic regression analysis showed Alb level (P = 0.02, adjusted OR [AOR] 0.29; 95% CI, 0.10–0.87) and TSF >15 mm (P = 0.01; AOR, 0.23; 95% CI, 0.07–0.78) were independent risk factors for SSI (see Table 3).

The NHSN⁶ reported 1.9% of patients underwent 1 of 39 types of surgery, including colon surgery. The JHAIS reported that 16.3% of patients who underwent colorectal surgery had a SSI.² The overall SSI rate in this study was similar to the JHAIS rate. The results of this study suggest TSF before surgery and serum Alb level after colorectal surgery may be associated with an increased risk of SSI. 

This study showed TSF is an independent risk factor for SSI; this is consistent with other studies. An American study¹⁶ reported TSF >22.8 mm is a risk factor for SSI. One Japanese study¹⁷ reported TSF >20 mm is a risk factor for SSI. Of the 24 patients with SSI in the current study, 16 had TSF >15 mm. It is interesting that the SGA scores of 12 of the 16 patients with SSI classified them as well nourished, and none of the patients had a score of 0 (ie, severe malnutrition). This might indicate that SGA score does not precisely predict a likelihood of SSI after colorectal surgery but that preoperative TSF, a noninvasive and easily calculable index, might help clinicians estimate SSI risk. Clinicians might calculate TSF once the surgery is planned and provide support to the patient with regard to losing weight before surgery; the authors have developed a preoperative weight-loss program. Comprehensive SSI risk assessment according to TSF, patient weight, and SGA score would be beneficial.

This study showed patients who developed SSI resumed oral dietary intake later than those who did not develop SSI. The Enhanced Recovery After Surgery Program²² recommends minimizing the fasting period and resuming oral intake within 24 hours unless contraindicated. Furthermore, a systematic review²³ showed prolonged fasting reduces intragestational immunity. The results of this study support these existing recommendations. 

Stoma placement minimally influences the anastomosis site, compared with upper digestive tract surgery. Guidelines²² support its safety and effectiveness.²⁴ Perioperative malnutrition worsens the likelihood of infection by affecting immunity. The colon provides a specific barrier and immune functions.²⁵ The results of this study suggest the potential of early resumption of oral dietary intake helps avoid SSI in elective colorectal surgery.²⁶

CDC guidelines⁸ do not include malnutrition as a predictor of SSI. However, these guidelines recommend the use of nutritional assessment parameters such as serum Alb level to screen for malnutrition.⁸ Serum Alb level should not be used as the only index because levels readily fluctuate owing to other factors such as intracellular fluid status and surgical invasion. The European Society for Clinical Nutrition and Metabolism²⁴ sets thresholds for requiring perioperative nutrition management, which includes an Alb level of >3.0 g/dL. A clinical study²⁷ suggests surgical invasion demands double the level of protein catabolism and increasing energy consumption. Improving Alb level after surgery is challenging, but a continuity of care that addresses the importance of a well-balanced diet and exercise and providing education to patients before surgery would be a potentially beneficial nursing intervention. A clinical trial regarding preoperative education is warranted.

This study has several clinically relevant limitations. The first involves the issue with generalizability because of the use of a convenience sampling method. Therefore, further studies with randomly selected patients are needed to confirm these results. In addition, because the half-life of serum Alb is 21 days, this index does not reflect immediate nutritional status. However, serum Alb is commonly examined in patients undergoing colorectal surgeries before surgery; thus, practitioners can conveniently evaluate the patient’s nutritional status.

A prospective cohort study involving 155 patients in 17 hospitals showed TSF >15 mm and low Alb levels are independent risk factors for SSI occurrence among patients undergoing colorectal surgeries. These study results confirm current recommendations to maximize the preoperative nutritional status of patients in order to prevent major complications and confirm that being overweight is a risk factor for SSI in patients undergoing colorectal surgery.

The authors acknowledge the nurses and surgeons of the 17 hospitals that participated in the data collection for this study.

Dr. Nakagawa is Assistant Professor, The Jikei University Graduate School of Medicine Program in Nursing, Kokuryocho Chofu, Tokyo. Ms. Ohno, Dr. Ikeda, and Dr. Muto are Professors, Graduate School, International University of Health and Welfare, Tokyo, Japan.

Please address correspondence to: Hiromi Nakagawa, Assistant Professor, The Jikei University Graduate School of Medicine Program in Nursing, 8-3 Kokuryocho, Chofu, Tokyo 182-8570; email: hiromin4@mail.goo.ne.jp.