Pre-Operative and Early Post-Operative Factors Associated with Surgical Site Infection after Laparoscopic Sleeve Gastrectomy

Abstract

Background:

Surgical procedures on obese patients are expected to have a high incidence of surgical site infection (SSI). The identification of pre-operative or early post-operative risk factors for SSI may help the surgeon to identify subjects in risk and adequately optimize their status. We conducted a study of the association of comorbidities and pre- and post-operative analytical variables with SSI following laparoscopic sleeve gastrectomy for the treatment of morbid obesity.

Patients and methods:

We performed a prospective study of all morbidly obese patients undergoing laparoscopic sleeve gastrectomy as a bariatric procedure between 2007 and 2011. An association of clinical and analytical variables with SSI was investigated.

Results:

The study included 40 patients with a mean pre-operative body mass index (BMI) of 51.2±7.9 kg/m². Surgical site infections appeared in three patients (7.5%), of whom two had an intra-abdominal abscess located in the left hypochondrium and the third had a superficial incisional SSI. Pre-operatively, a BMI >45 kg/m² (OR 8.7; p=0.008), restrictive disorders identified by pulmonary function tests (OR 10.0; p=0.012), a serum total protein concentration <5.3 g/dL (OR 13; p=0.003), a plasma cortisol >30 mcg/dL (OR 13.0; p=0.003), and a mean corpuscular volume (MCV) <82 fL (OR 1.6; p=0.04) were associated with post-operative SSI. Post-operatively, a serum glucose >128 mg/dL (OR 4.7; p=0.012) and hemoglobin <11g/dL (OR 7.5; p=0.002) were associated with SSI.

Conclusions:

The study supports the role of restrictive lung disorders and the values specified above for preoperative BMI, serum total protein and cortisol concentrations, and MCV, and of post-operative anemia and hyperglycemia as risk factors for SSI. In these situations, the surgeon must be aware of and seek to control these risk factors.

Obesity is a risk factor for surgical site infection (SSI) following various surgical procedures [1 –3]. Accordingly, surgical procedures performed exclusively on obese patients might be expected to have a high incidence of SSI. Some studies have determined that rates of SSI following bariatric surgery (BS) may be as high as 10% for open procedures and 3% for laparoscopic procedures [4]. The U.S. Centers for Disease Control and Prevention has published data on gastric operations (in which the rate of SSI ranges from 2.6% to 8.3%), but has not published data specifically for BS [5]. A multicenter study focused specifically on BS found lower rates of SSI of only 2.9% in patients who underwent any type of bariatric procedure and 0.8% in patients who specifically underwent laparoscopic BS [6].

Comorbidities associated with morbid obesity, such as diabetes mellitus, hyperlipidemia, heart disease, lung disease, hypertension, and insulin resistance play an essential role in the development of SSI [7,8]. Given that BS is not an emergency procedure, it is important to prepare the patient pre-operatively, with efforts to minimize the risk of all possible post-operative complications. The identification of pre-operative or early post-operative risk factors for SSI may help the surgeon to identify subjects at risk and adequately optimize their status.

The aim of the present study was to investigate the association of comorbidities and pre- and post-operative analytical variables with SSI after laparoscopic sleeve gastrectomy, on the possibility that this might permit the modification of these variables to minimize the post-operative risk of SSI.

Patients and Methods

The study was conducted in the Department of Surgery of the General University Hospital Elche as a prospective study of all morbidly obese patients undergoing laparoscopic sleeve gastrectomy as a bariatric procedure between October 2007 and May 2012.

Pre-operative evaluation

A multidisciplinary team including surgeons, endocrinologists, dieticians, endoscopists, radiologists, cardiologists, pulmonologists, anesthesiologists, psychiatrists, psychologists, and specialized nurses performed a combined medical, nutritional, and endocrinologic workup to evaluate potential candidates for BS. Pre-operative assessment included abdominal and cardiac ultrasonography, upper gastrointestinal endoscopy, pulmonary functional tests, and analytical evaluation of the candidates' hormonal and nutritional status. Psychiatrists and psychologists conducted additional interviews to evaluate the implications for the candidates of following a strict diet post-operatively. A dietician established a diet providing a total daily energy intake of 1,200 kCal, which was similar to the diet the candidates would have to follow post-operatively. Weight loss of at least 5% of the patient's excess weight was considered an indispensable condition for selection as a candidate for laparoscopic sleeve gastrectomy. Patients with documented gastroesophageal reflux disease were excluded.

Surgical technique

A laparoscopic approach was used in all patients. Five ports were established, at subxiphoideal and supraumbilical sites, in the right and left hypochondrium, and in the left flank. Short gastric vessels of the greater curvature of the stomach were divided with a harmonic scalpel (Ethicon Endo-Surgery, Somerville, NJ). A longitudinal resection from the angle of His to approximately 3–4 cm proximal to the pylorus was done with an endoscopic stapling device (Echelon Flex, Ethicon Endosurgery). The sleeve was calibrated prior to the resection with a 50-French bougie inserted along the lesser curve of the stomach. A Jackson–Pratt drain was inserted along the staple line of the sleeve and left in place.

Post-operative course

At 24 h after surgery, a blood analysis including the same parameters measured in the pre-operative sample was done and the patient was given an oral bolus of methylene blue dye. If dye was not present drain effluent, the patient began an oral intake of water or chamomile infusions. On post-operative day (POD) two the patient was allowed oral intake of high-protein liquid nutritional supplements to a total volune of 600 mL. Patients whose post-operative course was uneventful were usually on discharged on POD 3.

Variables

Complications and mortality were recorded. Clinical parameters, including pre-operative weight and BMI, comorbidities, and previous abdominal operations, were analyzed. The pre- and post-operative analytical samples were used for a complete blood count, clotting times, and serum biochemical assays of glucose, creatinine, albumin, triglycerides, cholesterol, transaminases, cortisol, parathormone, vitamin D, vitamin B₁₂, and folic acid concentrations.

Definitions

Surgical site infections were defined as superficial or deep incisional SSIs and intra-abdominal abscesses (organ/space SSIs). Incision SSI was defined as a purulent discharge from a surgical incision, as determined by an epidemiology nurse, and was confirmed with microbiologic culture. Surveillance for infection was continued for 30 d after discharge.

Intra-abdominal abscess was defined as a fluid collection identified by computed tomography in a symptomatic patient who presented with fever, abdominal pain, prolonged post-operative ileus, or sepsis. The diagnosis of intra-abdominal abscess was made by a radiologist. Evidence of staple-line leak was not considered to be SSI.

Statistical analyses

All statistical analyses were done with SPSS version 17.0 (SPSS Inc., Chicago, IL). A Gaussian distribution of variables was assessed with the Kolmogorov–Smirnoff test; values of p>0.05 were considered to have a Gaussian distribution. Quantitative variables following a normal distribution were defined by mean and standard deviation; non-Gaussian variables were defined by median and range. Qualitative parameters were defined by number of cases and percentages. Comparisons parameters were made with the Student t-test and Pearson correlation for quantitative parameters that had a Gaussian distribution, and with the Mann–Whitney U-test for variables with a non-Gaussian distribution. Comparison of qualitative parameters was done with the χ² method. Either linear or multi-variable regression analysis was used to determine the strength of association of independent parameters with outcome parameters. A value of p<0.05 was considered statistically significant.

Results

The study included 40 patients (38 females [95%] and two males [5%]), with a mean age of 43.2±10.2 years (range 20–62 years). Comorbidities included type 2 diabetes mellitus (T2DM) in 25% of the patients, dyslipidemia in 50% (40% hypercholesterolemia and 10% hypertrigliceridemia), hypertension in 30%, osteoarthritis in 20%, and obstructive sleep apnea syndrome in 16.6%. All of the patients with T2DM, hypertension, and dyslipidemia were being treated pharmacologically with acceptable control of their comorbidities. The patients' mean pre-operative BMI was 51.2±7.9 kg/m².

Surgical site infections occurred in three patients (7.5%), in two of whom as an intra-abdominal abscess located in the left hypochondrium and in the third as a superficial incisional SSI. The three patients with SSIs had uneventful post-operative courses and were discharged on the third day after surgery. The first two patients with SSIs presented to the emergency department complaining of fever over 38°C, on POD 5 and 6, respectively, and without any other clinical symptoms or signs on physical examination. Laboratory data for the first patient included a WBC of 18,000/mm³, a C-reactive protein (CRP) of 202 mg/L, and a fibrinogen concentration of 857 mg/dL; for the second patient the WBC was 16,800/mm³, CRP was 180 mg/L, and fibrinogen was 704 mg/dL, as the only remarkable findings. Computed tomographic scans enhanced with oral and intravenous contrast medium showed fluid collections of 5.5 cm and 3 cm, respectively, in the left hypochondrium of each patient, suggestive of an intra-abdominal abscess. A staple-line leak could not be demonstrated in any of the three cases of SSI. The patient with the 5.5-cm fluid collection underwent percutaneous drainage, which yielded a sauguina-purulent discharge that on culture grew Escherichia coli. The other patient with an abscess did not require drainage. In both cases antibiotic treatment was started with piperacillin-tazobactam at 4.5 g q6h maintained for 7 d. Both patients recovered satisfactorily.

The third patient who developed an SSI came to the emergency department on the fifth day after surgery, complaining of purulent discharge through a 2.5-cm incision in the patient's right hypochondrium, through which a culture specimen was extracted. The incision was opened and the purulent discharge was drained. A Penrose drain was left in place for 2 d. Systemic antibiotic therapy was not necessary. The patient recovered uneventfully with proper healing of the wound. Culture of the drained material grew E. coli.

Pre-operative variables associated with surgical site infection

The study patients' pre-operative BMI was associated with the development of post-operative SSI. The mean BMI of patients who did not develop an SSI post-operatively was 43.3±2.8 kg/m² and that of those who did was 47.2±4.8 kg/m² (p=0.007). A cutoff value of 45 kg/m² was established as the most discriminatory value for BMI as a risk factor for SSI (OR 8.7; 95% CI 1.5–12.3; p=0.008).

The finding of restrictive disorders in pulmonary function tests was also associated with SSI (OR 10; 95% CI 2.7–37.1; p=0.012). A significant association between SSI and forced expiratory volume in 1 second (FEV₁), forced vital capacity (FVC), of the Tiffeneau index (FEV₁/FVC) could not be established.

A significant association was found between pre-operative total protein concentration and the occurrence of SSI. The mean preoperative total protein concentration was 6.8±0.6 g/dL in patients who did not develop an SSI and 5.1±0.3 g/dL in those who did (p=0.001). A cutoff value of 5.3 g/dL (OR 13; 95% CI 2, 83.3; p=0.003) was established for mean total protein concentration as a risk factor for post-operative SSI.

The pre-operative serum cortisol concentration showed a significant association with post-operative SSI. The median serum cortisol concentration of patients who did not develop a post-operative SSI was 14 mcg/dL (range 2–27 mcg/dL) versus 66 mcg/dL (range 30–75) in those who did (p=0.006). A cutoff value of 30 mcg/dL (OR 17.9; 95% CI 2.7, 46; p=0.016) was established for serum cortisol concentration as a risk factor for SSI.

A significant association was also established between SSI and pre-operative mean corpuscular volume (MCV), which was 86.6±3.7 fL in patients without post-operative SSIs as opposed to 81.6±2.3 fL in those with post-operative SSIs (p=0.039). A cutoff value of 82 fL (OR 1.6; 95% CI 1.1–2.7; p=0.04) was established for MCV as a risk factor for SSI. Given the association between MCV and anemia, we investigated the study patients' hemoglobin concentrations. A trend was observed for the association of pre-operative hemoglobin levels with SSI (p=0.095).

A multivariable analysis was done with all parameters associated significantly with SSI by univariate analyses, but none of them reached the level of significant association because of the small sample size.

Post-operative variables associated with surgical site infection

With regard to association of the post-operative serum glucose concentration with SSI, the mean serum glucose concentration of patients without post-operative SSIs was 102.5±18.9 mg/dL, versus 149±26.9 mg/dL for those who had post-operative SSIs (p=0.012). A cutoff value of 128 mg/dL (OR 4.7; 95% CI 2.0–12.7; p=0.012) was established for postoperative serum glucose concentration as a risk factor for SSI.

A significant association was also established between SSI and post-operative hemoglobin concentrations. The mean hemoglobin of patients without post-operative SSIs was 11.9±1 g/dL, as opposed to 10.6±1.8 g/dL for patients who developed post-operative SSIs (p=0.005). A cutoff value of 11 g/dL (OR 7.5; 95% CI 2.1, 27; p=0.002) was established for hemoglobin as a risk factor for SSI. Given its association with SSI, we also investigated post-operative MCV, observing a trend toward association (p=0.07).

A multivariable analysis including post-operative glucose and hemoglobin levels did not show either of these variables as reaching a significant association with SSI because of the small sample size. Pre- and post-operative risk factors for SSI as identified in the study are summarized in Table 1.

Table 1.

Pre- and Early Post-Operative Risk Factors for Surgical Site Infection After Laparoscopic Sleeve Gastrectomy

Risk factor	Odds ratio
Pre-operative
Body mass index >45 kg/m²	8.7
Restrictive lung disorders	10
Total protein <5.3 g/dL	13
Cortisol >30 mcg/dL	17.9
Mean corpuscular volume (MCV) <82 fL	1.6
Early post-operative
Hemoglobin <11 g/dL	7.5
Glucose >128 mg/dL	4.7

Discussion

Patients with a BMI >30 kg/m² are more prone to develop SSI and pneumonia than patients with lower values of BMI [9]. Among morbidly obese patients undergoing BS, a cutoff point of BMI for increased susceptibility of SSI has not been established. In our series the cutoff point was a BMI of 45 kg/m². Patients with a pre-operative BMI exceeding this had an 8.7-fold higher risk of developing a post-operative SSI than those with a BMI below it. It is advisable that these patients follow a diet pre-operatively because this facilitates BS by reducing liver volume and intra-abdominal adipose tissue, and reduces the risk of pulmonary and thrombotic complications [10]. This is especially important in patients who will undergo a restrictive technique of BS, so that their pre-operative weight loss will reflect their post-operative loss and increase the likelihood of success of the surgical treatment [11]. Although many patients cannot be expected to achieve a pre-operative BMI <45 kg/m², some reduction in BMI could substantially reduce the risk of post-operative SSI.

Because severe obesity is often associated with defective pulmonary function, and abdominal surgery increases the risk of respiratory post-operative complications (RPC), an increased incidence of RPC might follow BS. Respiratory postoperative complications are associated with increased mortality. The most frequent defect in pulmonary function is a restrictive disorder, because the increase of intra-abdominal volume with BS impairs proper expansion of the lungs during the inspiration [12]. This alteration in pulmonary function leads to hypoxemia and a decrease in tissue oxygen tension. Oxygen is essential for the bacterial lysis produced by neutrophils and macrophages [13], and in our series, morbidly obese patients with restrictive pulmonary disorders had a 10-fold increased risk of SSI. Pre-operative weight loss and respiratory physiotherapy may help to improve these patients' restrictive pulmonary status and reduce their risk of SSI.

The serum total protein concentration is considered a marker of nutritional status, and in patients undergoing major surgery is a prognostic indicator [14]. Hypoproteinemia leads to impaired immune function, increasing the propensity of such patients to SSI [15]. Moreover, hypoproteinemia in patients undergoing BS may also increase the risk of staple line or anastomotic leaks, as happens with surgery in other tissues [16]. In our series, pre-operative protein concentrations under 5.3 g/dL were accompanied by a 13-fold higher risk of post-operative SSI. This can be easily controlled by starting oral nutritional high-protein supplements at least 2 wks before surgery.

Mild hypercortisolism has been associated with metabolic syndrome and morbid obesity [17]. Cortisol appears to play a role in adiposity, especially in central obesity [18,19], but is not correlated with BMI [20], which was confirmed in our patients. On the other hand, hypercortisolemia induces a state of immunocompromise that predisposes the obese patient to SSI [21]. In our series, cortisol levels over 30 mcg/dL were associated with a 17.9-fold higher risk of developing SSI. Given that hypercortisolism is related to adiposity, pre-operative weight loss may be the best chance to control this risk factor. Moreover, in the follow-up of our patients presenting with increased serum cortisol concentrations, these fell into the normal range of progressive post-operative weight loss.

In our series, a decreased pre-operative MCV and early post-operative anemia were associated with a higher risk of SSI. Mean corpuscular volume and hemoglobin concentrations are parameters strongly linked to one another. Because of the small sample size in our study, pre-operative hemoglobin concentrations and post-operative values of MCV did not reach statistical significance as risk factors for SSI, but both showed a trend toward this. Anemia has been associated with the nutritional status of patients, and, as explained earlier, leads to a decline in the immune status [15]. On the other hand, the reduced oxygen tension in anemic patients impairs the lytic function of phagocytes, favoring the development of SSI [13]. Given this, it can be postulated that blood transfusion or the administration of recombinant erythropoietin to any anemic patient before surgery, and the transfusion of red blood cell-concentrates to any patient presenting with a hemoglobin concentration <11 g/dL on the first post-operative day, may reduce the risk of SSI in BS. Surgical site infections include superficial incisional SSIs that cannot be considered severe post-operative complications, and the use of blood transfusion to avoid infection may therefore be unwarranted. More studies, with larger numbers of patients, must be conducted to assess these suggestions.

Early post-operative hyperglycemia was also associated with SSI in our series. Higher peri-operative serum glucose concentrations are associated with complications after various types of surgery, including delays in wound healing, a higher risk of SSI, cardiac complications, and increased mortality rates [22,23]. In hyperglycemic situations, concentrations of catecholamines, corticosteroids, and growth hormone are increased, all of which inhibit oxygen release and favor SSI [24]. In our series, a cutoff point for serum glucose could be established at 128 mg/dL, with patients whose glucose exceeded this having a 4.7-fold higher risk of SSI. This finding reinforces the rationale for aggressive insulin treatment during the first several post-operative days for patients who have post-operative hyperglycemia. However, the important risk of hypoglycemia with this treatment must be kept in mind, and close monitoring of patients' glycemic status is mandatory, especially for diabetic patients [13,25 –28].

The two main limitations of this study were its small sample size and the predominance of females in the study population. Future studies, with greater numbers of patients, may confirm the results we obtained. The small sample size did not allow us to determine which of the factors we analyzed were independently associated with SSI by multivariable analysis. Additionally, pre-operative anemia showed a trend toward an association of SSI with pre-operative hemoglobin concentrations (p=0.095). With regard to the female predominance in our population sample, sleeve gastrectomy is an indicated procedure only for select patients, and a weight loss of at least 5% of the patient's excess weight was considered an indispensable condition so that patients' pre-operative weight loss will reflect their post-operative loss [11]. This requirement has in our experience yielded excellent results in post-operative weight loss, and female patients have shown better adaptation to such a pre-operative diet, which was the main reason of the female predominance in our study population.

This study supports the role of pre-operative BMI, restrictive lung disorders, serum total protein, cortisol concentration and MCV, and post-operative anemia and hyperglycemia as risk factors for developing SSI. In these situations, the surgeon must take action to control some modifiable risk factors.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Olsen

, Nepple

, Riew

et al. Risk factors for surgical site infection following orthopaedic spinal operations. J Bone Joint Surg Am, 2008; 90:62–69.

Dindo

, Muller

, Weber

et al. Obesity in general elective surgery. Lancet, 2003; 361:2183–2189.

Flancbaum

, Choban

. Surgical implications of obesity. Ann Rev Med, 1998; 29:215–234.

Nguyen

, Goldman

, Rosenquist

et al. Laparoscopic versus open gastric bypass: A randomized study of outcomes, quality of life and costs. Ann Surg, 2001; 234:279–289.

Christou

, Jarand

, Sylvestre

et al. Analysis of the incidence and risk factors for wound infections in open bariatric surgery. Obes Surg, 2004; 14:16–22.

The Longitudinal Assessment of Bariatric Surgery (LABS) Consortium. Perioperative safety in the longitudinal assessment of bariatric surgery. N Engl J Med, 2009; 361:445–454.

Freeman

, Anderson

, Hartwig

et al.

Surgical site infections following bariatric surgery in community hospitals: A weighty concern?

Obes Surg, 2011; 21:836–840.

Woeste

, Muller

, Bechstein

, Wullstein

. Increased serum levels of C-reactive protein precede anastomotic leakage in colorectal surgery. World J Surg, 2010; 34:140–146.

Dossett

, Dageforde

, Swenson

et al. Obesity and site-specific nosocomial infection risk in the intensive care unit. Surg Infect, 2009; 10:137–142.

10.

Buchwald

, Williams

. Bariatric surgery worldwide 2003. Obes Surg, 2004; 14:1157–1164.

11.

Martinez-Ramos

, Salvador-Sanchis

, Escrig-Sos

. Pérdida de peso preoperatoria en pacientes candidatos a cirugía bariátrica. Recomendaciones basadas en la evidencia. Cir Esp, 2012; 90:147–155.

12.

Farina

, Crimi

, Accogli

et al. Preoperative assessment of respiratory function in severely obese patients undergoing biliopancreatic diversion. Eur Surg Res, 2012; 48:106–110.

13.

Wesley Alexander

, Solomkin

, Edwards

. Updated recommendations for control of surgical site infections. Ann Surg, 2011; 253:1082–1093.

14.

Gibbs

, Cull

, Henderson

et al. Preoperative serum albumin level as a predictor of operative mortality and morbidity: Results from the National VA Surgical Risk Study. Arch Surg, 1999; 134:36–42.

15.

Rham

. A guide to perioperative nutrition. Aesthetic Surg J, 2004; 24:385–390.

16.

Ryan

, Hearty

, Prichard

et al. Association of hypoalbuminemia on the first postoperative day and complications following esophagectomy. J Gastrointest Surg, 2007; 11:1355–1360.

17.

Anagnostis

, Athyros

, Tzionalos

et al. The pathogenic role of cortisol in the metabolic síndrome: A hypothesis. J Clin Endocrinol Metab, 2009; 94:692–701.

18.

Rebuffe-Scrive

, Walsh

, McEwen

et al. Effect of chronic stress and exogenous glucocorticoids on regional fat distribution and metabolism. Physiol Behav, 1992; 67:1122–1128.

19.

Obunai

, Jani

, Dangas

. Cardiovascular morbidity and mortality of themetabolic syndrome. Med Clin North Am, 2007; 91:1169–1184.

20.

Holt

, Wild

, Postle

et al. Cortisol clearance and associations with insulin sensivity, body fat and fatty liver in middle-aged men. Diabetologia, 2007; 50:1024–1032.

21.

Fareau

, Vassilopoulou-Sellin

. Hypercortisoleia and infection. Infect Dis Clin North Am, 2007; 21:639–657.

22.

Furnary

, Wu

, Bookin

. Effect of hyperglycemia and continuous intravenous insulin infusions on outcomes of cardiac surgical procedures: The Portland Diabetic Project. Endocr Pract, 2004; 10:21–33.

23.

Van Den Berghe

, Wouters

, Weekers

et al. Intensive insulin therapy in critically ill patients. N Engl J Med, 2001; 345:1359–1367.

24.

Turina

, Fry

, Polk

Jr.

Acute hyperglycemia and the innate immune system: Clinical, cellular, and molecular aspects. Crit Care Med, 2005; 33:1624–1633.

25.

Kramer

, Groom

, Weldner

et al. Glycemic control and reduction of deep sterna wound infection rates: A multidisciplinary approach. Arch Surg, 2008; 143:451–456.

26.

Hruska

, Smith

, Hendy

et al. Continuous insulin infusion reduces infectious complications in diabetics following coronary surgery. J Cardiac Surg, 2005; 20:403–407.

27.

Swenne

, Lindholm

, Borowiec

et al. Peri-operative glucose control and development of surgical wound infections inpatients undergoing coronary artery bypass graft. J Hosp Infect, 2005; 61:201–212.

28.

Ramos

, Khalpey

, Lipsitz

et al. Relationship of perioperative hyperglycemia and postoperative infections in patients who undergo general and vascular surgery. Ann Surg, 2008; 248:585–591.