Are Adiponectin and Leptin Good Predictors of Surgical Infection after Colorectal Surgery? A Prospective Study

Abstract

Background:

Infections are the most frequent complication after colorectal surgery. It has been suggested that adipose tissue metabolism could be related to the risk of post-operative infection, but this could be partially related to the body-mass index. The aim of this study was to look for a relation between adipocytokine levels and the risk of post-operative infection and its type.

Methods:

This prospective cohort study was conducted between March 2013 and March 2014 in two French teaching hospitals. Pre-operative plasma levels of adiponectin and leptin were measured in consecutive patients undergoing elective colorectal surgery. All infections in the 30 d following surgery were recorded.

Results:

Among the 142 patients included, 29 (20.4%) presented a post-operative infection: 26 surgical site infections and three extra-abdominal infections. Adiponectin and leptin levels correlated weakly but substantially with the body mass index (r_spearman=−0.26 and +0.31, respectively). While there was no substantial difference between patients with and those without post-operative infection for adiponectin, median pre-operative leptin was substantially greater in patients with post-operative infection (8.67 vs. 4.37 ng/mL, p=0.003). A substantial interaction was found between leptin and cancer. In patients operated on for cancer, the area under the receiver-operating characteristic (ROC) curve was lower than in patients with benign diseases (0.597 vs. 0.858, p=0.011). Similar results were observed for intra-abdominal infection and surgical site infection.

Conclusion:

Patients with greater levels of leptin before colorectal surgery have an increased risk of post-operative surgical infection. This effect is stronger in patients without cancer. Adiponectin levels are not related to the risk of infection in Western patients.

Post-operative infections are the most frequent complications after colorectal surgery. They may be life threatening, prolong hospitalization, increase the cost of care, and adversely affect the cancer outcome [1,2]. Wound infection is the most frequent complication after colorectal surgery with an incidence around 20%, and even greater in obese patients [3 –9]. Adipose tissue produces several adipocytokines involved in inflammation and insulin resistance, two capital processes in current understanding of infection. It has been suggested that the adipocytokines adiponectin and leptin could be predictors of the risk of surgical site infection (SSI) [10,11]. Current understanding of their precise physiological role is limited. It seems that although adiponectin has anti-inflammatory properties, leptin acts as an acute-phase reactant that modulates the inflammatory response [12,13]. However, although the levels of both adipocytokines are related to the total amount of fatty tissue, obese patients do not seem to carry an increased risk of post-operative infections other than SSI [7,8]. With the aim to understand the role of adipose tissue metabolism and inflammation in the onset of surgical infection, previous works have measured the pre-operative levels of different adipocytokines and inflammatory markers in patients undergoing elective colorectal surgery.

The relation between leptin and adiponectin levels and the onset of post-operative infection has been investigated, mainly in studies that included a small number of patients, with conflicting results [10,11,14]. The aims of this study were to investigate the relation between the pre-operative plasma adipocytokine levels and the risk of post-operative infection, and to evaluate adiponectin and leptin as predictors of post-operative infection.

Patients and Methods

Study design and inclusion criteria

This bi-institutional prospective observational study was conducted between March 2013 and March 2014 at the Bocage Central University Hospital and the George-François Leclerc Anticancer Centre, both in Dijon (France). All consecutive patients undergoing elective colorectal surgery with anastomosis in either of the two participating institutions were included, whatever the diagnosis (benign or malignant disease). Exclusion criteria were the association of hyperthermic intra-peritoneal chemotherapy for peritoneal carcinomatosis and the presence of any sign of intra-abdominal infection before or during surgery. The protocol was approved by the regional Research Ethics Committee (CPP Est 1, Dijon, France) and the French Agency for the Security of Health Products (AFSSAPS). Patients provided written informed consent before their inclusion in the study.

Endpoints

The two endpoints of the study were the overall frequency of post-operative infections and the specific incidence of SSI and intra-abdominal infections in the 30 d following the operation. The definitions used for intra-abdominal infection, SSI infection, pneumonia, urinary-tract infection, and central-catheter infection were those of the U.S. Centers for Disease Control and Prevention [15].

Evaluation of the patients and follow-up

Potential patient-specific and intra-operative risk factors for infection were recorded (gender, age, pre-operative health status, body-mass index [BMI] and surgical indication). The afternoon before surgery, a blood sample was taken to determine the pre-operative plasma levels of C-reactive protein (CRP), leptin, and adiponectin. Pre-operative intravenous antibiotics were administered before the skin incision according to current national French guidelines and a consensus statement. During the post-operative period, patients were examined by the attending surgeon daily. He/she was blinded to the adipocytokine results. Patients were observed at the outpatient clinic within 5 wks after surgery and all complications were recorded.

Biological analysis

Blood samples were separated immediately by centrifugation and stored at –80°C until centralized analysis. Serum C-reactive protein (CRP) was measured using conventional laboratory techniques. Plasma concentrations of leptin (ng/mL) and adiponectin (mcg/mL) were assayed using a commercially available enzyme-linked immunosorbent assay (ELISA) kit (Quantikine; R and D Systems Inc., Minneapolis, MN, USA) according to the manufacturer's instructions. The intra- and inter-assay coefficients of variation were below 5% for all measurements.

Statistical analysis

Categorical variables were expressed as percentages and compared with the Chi-2 test. As the Shapiro-Wilk test was substantial for all variables, quantities were expressed as medians with interquartile ranges (IQR) and compared using non-parametric tests (Mann-Whitney or Kruskal-Wallis, as appropriate). Correlations between each adipocytokine and BMI were measured using the Spearman correlation coefficient (r_s). Odds ratios (ORs) and their 95% confidence intervals (95% CI) were calculated using logistic regression models to assess the association between each adipocytokine and the risk of infection (post-operative infection, intra-abdominal infection, and SSI). A transformation with base 2 logarithm was performed aiming that the OR showed the impact of doubling adipocytokines' value on the risk of infection. Log-linearity was checked using fractional polynomials. In addition to the BMI, potential confounders in multivariate analyses were all clinical and biological variables related to post-operative infections or to plasma adipocytokines with a p value<0.20 in univariate analysis. Potential interactions between the recorded clinical data and the impact of adipocytokines were also assessed. The discriminant ability of substantial adipocytokines was assessed with the area under the ROC curve, which is presented with its 95% confidence interval. The statistical analysis was performed using Stata software (Stata corporation, College Station TX, USA). The significance level was p<0.05 for all tests.

Results

Description of patients and procedures

One hundred forty-two patients (78 men and 64 women) with a mean age of 66.3±13.7 years were included in the study; 97 of these were operated on for colorectal cancer and 45 for benign diseases. The 30-d mortality rate was 2.8% (four patients).

Post-operative infections

Twenty-nine patients (20.4%) developed a post-operative infection. Among these, 26 had a SSI (overall incidence of 18.3%): 12 had a wound infection (8.5%) and 15 had an intra-abdominal infection (10.6%), with one patient presenting both. Three patients had only extra-abdominal infections: Urinary tract infections (n=2), and catheter infection (n=1). Two patients had both an SSI and an extra-abdominal infection.

Factors related to adipocytokine levels

The pre-operative concentrations of adipocytokines correlated with the age, gender and BMI (Table 1). Namely, pre-operative plasma adiponectin correlated weakly but substantially with BMI (negative correlation, r_s=−0.26, p=0.003) and with age (positive correlation, r_s=0.19, p=0.028). It was also greater in females than in males (9.98 vs. 4.87 mcg/mL, p=0.003). The same associations were found for leptin, except for BMI (positive correlation, rs=0.31, p=0.0003).

Table 1.

Plasma Adipocytokine Levels According to the Clinical Data. Medians and Interquartile Intervals (Q1–Q3) Are Presented

	Leptin (ng/mL)			Adiponectin (mcg/mL)
	Median	Q1–Q3	p	Median	Q1–Q3	p
Gender			0.0003			0.003
Females	6.29	3.53–15.16		9.98	4.89–16.50
Males	3.73	2.45–7.44		4.87	2.61–12.89
Age			0.018			0.129
<68	3.77	2.50–7.38		5.80	3.06–12.89
≥68	5.83	3.09–10.43		9.14	4.19–15.97
BMI			0.020			0.024
<25	3.77	2.50–6.46		10.29	3.84–20.30
25–30	5.12	3.01–8.96		6.09	4.16–12.48
≥30	9.53	3.30–36.1		4.93	1.83–9.29
Cancer			0.680			0.571
Yes	4.83	2.72–9.69		6.79	3.65–13.06
No	4.70	2.71–8.52		7.67	3.76–15.44

Q1–Q3=Interquartile intervals.

The correlation coefficient between leptin and age was 0.17 (p=0.043), and leptin was also greater in females than in males (6.28 vs. 3.73 ng/mL, p=0.002). There was no correlation between adiponectin and leptin (r_s=-0.03, p=0.682). Neither adiponectin nor leptin was associated with pre-operative CRP or with the surgical indication.

Overall risk of infection and adipocytokine levels

Pre-operative plasma leptin was the only adipocytokine substantially associated with post-operative infections, intra-abdominal infections, or surgical site infections (Table 2). After adjustment for the interaction terms and for BMI, age, gender and cancer, the adjusted leptin OR associated with post-operative infection (reflecting the impact on the risk of infection after doubling the leptin values) was 4.66 (95% CI: 1.92–11.28), p=0.001. Consistent results were observed for SSI: adjusted OR=6.46 (2.25–18.57), p=0.001; and also for intra-abdominal infections: adjusted OR=4.73 (1.76–12.72), p=0.002. A substantial interaction between leptin and cancer was found for the risk of post-operative infection (p=0.022). In patients operated on for cancer the area under the ROC curve, assessing the ability of leptin to discriminate between patients with or without infection was substantially lower than in patients with benign diseases for all types of infection (Table 3).

Table 2.

Plasma Adipocytokine Levels and Other Laboratory and Clinical Data According to the Presence of Post-Operative Infection, Intra-Abdominal Infection and Surgical Site Infection. Medians and Interquartile Intervals (Q1–Q3) Are Presented

Overall Post-operative Infection
	No infection (n=113)	Post-operative infection (n=29)	p
Age (years)	66 (59–77)	72 (64–77)	0.295
Gender			0.990
Males	54.9%	55.2%
Females	45.1%	44.8%
Body-mass index (kg/m²)	25.6 (22.5–28)	25.5 (22.9–29.9)	0.772
Leptin (ng/mL)	4.37 (2.54–7.60)	8.67 (4.08–15.14)	0.003
Adiponectin (mcg/mL)	7.12 (3.76–13.95)	6.79 (3.56–14.54)	0.808
C reactive protein (mg/L)	4 (3–13)	4 (3–16)	0.903
Intra-Abdominal Infection
	No intra-abdominal infection (n=127)	Intra-abdominal infection (n=15)	p
Age (years)	68 (59–77)	70 (57–74)	0.722
Gender			0.676
Males	54.3%	60%
Females	45.7%	40%
Body-mass index (kg/m²)	25.8 (22.9–28.4)	24.1 (20.8–27.4)	0.297
Leptin (ng/mL)	4.51 (2.55–8.87)	8.67 (4.08–12.82)	0.025
Adiponectin (mcg/mL)	6.82 (3.76–13.59)	10.76 (2.33–19.33)	0.483
C reactive protein (mg/L)	5 (3–15)	3 (3–6)	0.097
Surgical Site Infection
	No surgical site infection (n=116)	Surgical site infection (n=26)	p
Age (years)	66.5 (58.5–77)	71.5 (64–78)	0.319
Gender			0.754
Males	54.3%	57.7%
Females	45.7%	42.3%
Body-mass index (kg/m²)	25.6 (22.7–28)	25.4 (22.9–29.9)	0.933
Leptin (ng/mL)	4.46 (2.55–7.94)	8.83 (3.77–15.14)	0.006
Adiponectin (mcg/mL)	7.07 (3.84–13.80)	7.33 (3.02–15.04)	0.857
C reactive protein (mg/L)	4.5 (3–13.9)	3.5 (3–11)	0.631

QI=Interquartile intervals.

Table 3.

Accuracy of Leptin (Assessed by the Area Under the ROC Curve) To Predict the Risk of Infection in Patients with and without Cancer

	Cancer patients (n=97)		No cancer patients (n=45)
Type of infection	AUC	95% CI	AUC	95% CI	p
Post-operative infection	0.597	[0.493–0.696]	0.858	[0.732–0.949]	0.011
Surgical site infection	0.574	[0.473–0.677]	0.895	[0.759–0.963]	0.002
Intra-abdominal infection	0.572	[0.473–0.677]	0.885	[0.759–0.963]	0.003

AUC=Area Under the ROC Curve; CI=Confidence interval; ROC=receiver-operating characteristic.

Discussion

Adipocytokines reflect the metabolic activity of the adipose tissue. The relation between obesity, insulin-resistance, inflammation, and infection is a major concern in the current literature dealing with surgical infection. Given the disparity of results in recent studies, our aim was to provide more data to understand the relation between adipocytokines and the risk of surgical infection.

Two Japanese groups studied adipocytokines as predictors of surgical infection, and concluded that adiponectin was a better predictor than leptin. Matsuda et al. measured both adipocytokines in 41 patients undergoing colorectal cancer surgery [10]. In their univariate analysis, pre-operative concentrations of both adipocytokines were substantially lower in patients who subsequently developed post-operative infection. Only adiponectin remained substantial in the multivariate analysis. After having assessed adiponectin in animal models of sepsis, Yamamoto et al. focused on adiponectin alone in their series of 150 patients undergoing gastric cancer surgery [14]. They found a link between high pre-operative concentrations of adiponectin and the overall risk of infection. They also stated that the post-operative decrease in adiponectin might be involved in the onset of infection. The results of the AGARIC study, which included 174 patients undergoing elective colorectal surgery, found quite different results: Greater pre-operative concentrations of leptin, CRP, and insulin in infected patients, and no substantial relation between adiponectin and post-operative infection. Six other adipocytokines measured in this study were unrelated to the onset of post-operative infection [11].

Our results confirmed that patients with greater pre-operative concentrations of leptin have an increased risk of post-operative infection as a whole, and of SSI and intra-abdominal infection in particular. This is consistent with the results of our previous work on this topic and confirms that leptin is a marker of the risk of surgical infection [11]. Whether it has or not a physiologic role in the onset of infection remains unknown. Leptin is related to obesity, insulin resistance, and inflammation, all of which are involved in the onset of infection. It is an important signaling molecule in the regulation of food intake and energy balance, and it also acts as an acute-phase reactant that regulates pro-inflammatory immune responses.

It has been suggested that leptin deficiency may increase susceptibility to bacterial infections and hypothesized that it could impair or down regulate the inflammatory response [16]. Its precise physiologic role remains unknown, but it could be a marker of pre-operative ongoing inflammation, just as pre-operative C-reactive protein has been linked to an increased risk of post-operative infection [17,18]. As leptin is produced mainly in adipose tissue, its concentrations are partially related to the BMI, and particularly to the distribution of body fat in obese patients [11,19]. The BMI itself is a well-known risk factor of SSI, namely wound abscess, although this was not confirmed in the present study whose size number was not large enough to reach significance on that point [7 –9,20]. However, the present study showed that the correlation between leptin and infection goes beyond the effect of the BMI and the BMI-associated increased risk of wound abscesses. This would also be consistent with increasing evidence showing that though obesity increases the risk of wound abscess, it has no impact on the incidence of other infectious complications. In particular, it does not increase the risk of intra-abdominal or respiratory infection [6, 20 –23].

Interestingly, according to our results, the predictive accuracy of leptin for the onset of infection was much better in the absence of cancer. This explains probably a great part of the disparities found between our results and those obtained by the Japanese authors, as they focused only on cancer patients. As cancer patients have already an ongoing inflammatory activity induced by the neoplasm, the regulatory role of leptin in the inflammatory response may be impaired in that setting.

Adiponectin seems to have anti-inflammatory properties [13]. It has also been associated with insulin resistance and several cancers (breast, endometrium, colon, stomach, and prostate) [24]. As confirmed in the present study, circulating concentrations of adiponectin correlate inversely with the BMI. In this study, as in the AGARIC study, adiponectin was not substantially related to the risk of infection [11]. The two Japanese groups that studied this point found a relation but in opposite senses. For Yamamoto et al. infected patients presented greater concentrations, whereas for Matsuda et al. they had lower concentrations [10,14]. The hypothesis we can put forward to explain this discrepancy is ethnic and weight differences between Asiatic and European patients, even if we adjusted by the BMI and no interaction was found for it.

We therefore conclude that pre-operative leptin is related to the risk of surgical infection, particularly in the absence of cancer. Adiponectin does not seem to have such a relation in Western patients. Further studies on the physiological effects of leptin could help our comprehension of the relation between adipose tissue metabolism, inflammation, and the risk of surgical infection.

Footnotes

Author Disclosure Statement

No conflict of interest to declare.

This work was funded by a grant from the Groupement de Coopération Sanitaire Grand-Est 2011, the University Hospital of Dijon (France) and a French Government grant managed by the French National Research Agency under the program “Investissements d'Avenir” (ANR-11-LABX-0021).

Acknowledgments

The authors are indebted to Elisabeth Devilliers, MD, for her heartful support of this project. The authors thank the CIC-EC (CIC 1432), INSERM 866 and DRCI for their logistic support with special thanks to Emilie Galizzi, Cassandra Porebski, Amandine Martin, Fanny Lachaux, Evelyne Phu, and Maud Carpentier. The authors also thank the medical doctors and residents who took care of the patients: Giovanni Di Giacomo, Nicolas Lagoutte, Sophie Al Samman, Cédric Angot, Sophie Jambet, Sabine Holl, Pierre Landreau, Matthieu Poussier, Alexandre Doussot, Christophe Combier, Nicolas Santucci, Baptiste Borraccino, Thomas Perrin, Laura Vincent, and Axel Gilbert; the nurses who took the blood samples; the patients who accepted to participate in this protocol; the laboratory technicians who helped with the assays; and Mr Philip Bastable, who revised the manuscript.

References

Brown

, Mathew

, Keding

, et al. The impact of postoperative complications on long-term quality of life after curative colorectal cancer surgery. Ann Surg, 2014; 259:916–923.

Alves

, Panis

, Mathieu

, et al. Postoperative mortality and morbidity in French patients undergoing colorectal surgery. Arch Surg, 2005; 140:278–283.

Huebner

, Hübner

, Cima

, Larson

. Timing of complications and length of stay after rectal cancer surgery. J Am Coll Surg, 2014; 218:914–919.

Mallol

, Sabaté

, Kreisler

, et al. Incidencia de la infección de la herida quirúrgica en cirugía colorrectal electiva y su relación con factores perioperatorios. Cir Esp, 2012; 90:376–381.

Pujol

, Limón

, López-Contreras

, et al; VINCat Program. Surveillance of surgical site infections in elective colorectal surgery. Results of the VINCat Program (2007–2010). Enferm Infecc Microbiol Clin, 2012; 30:20–25.

Dindo

, Muller

, Weber

, Clavien

. Obesity in general elective surgery. Lancet, 2003; 361:2032–2035.

Hourigan

. Impact of obesity on surgical site infection in colon and rectal surgery. Clin Colon Rectal Surg, 2011; 24:283–290.

Makino

, Shukla

, Rubino

, Milsom

. The impact of obesity on perioperative outcomes after laparoscopic colorectal resection. Ann Surg, 2012; 255:228–236.

Balentine

, Wilks

, Robinson

, et al. Obesity increases wound complications in rectal cancer surgery. J Surg Res, 2010; 163:35–39.

10.

Matsuda

, Matsutani

, Sasajima

, et al. Preoperative plasma adiponectin level is a risk factor for postoperative infection following colorectal cancer surgery. J Surg Res, 2009; 157:227–234.

11.

Ortega-Deballon

, Duvillard

, Scherrer

, et al; AGARIC study group. Preoperative adipocytokines as a predictor of surgical infection after colorectal surgery: A prospective survey. Int J Colorectal Dis, 2014; 29:23–29.

12.

Ouchi

, Kihara

, Funahashi

, Nakamura

, et al. Reciprocal association of C-reactive protein with adiponectin in blood stream and adipose tissue. Circulation, 2003; 107:671–674.

13.

Yokota

, Oritani

, Takahashi

, et al. Adiponectin, a new member of the family of soluble defense collagens, negatively regulates the growth of myelomonocytic progenitors and the functions of macrophages. Blood, 2000; 96:1723–1732.

14.

Yamamoto

, Maeda

, Uji

, et al. Association between reduction of plasma adiponectin levels and risk of bacterial infection after gastric cancer surgery. PLoS One, 2013; 8(3):e56129.

15.

Horan

, Andrus

, Dudeck

. CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control, 2008; 36:309–332.

16.

Faggioni

, Moser

, Feingold

, Grunfeld

. Reduced leptin levels in starvation increase susceptibility to endotoxic shock. Am J Pathol, 2000; 156:1781–1787.

17.

Moyes

, Leitch

, McKee

, et al. Preoperative systemic inflammation predicts postoperative infectious complications in patients undergoing curative resection for colorectal cancer. Br J Cancer, 2009; 100:1236–1239.

18.

Ortega-Deballon

, Radais

, Facy

, et al. C-reactive protein is an early predictor of septic complications after elective colorectal surgery. World J Surg, 2010; 34:808–814.

19.

Minocci

, Savia

, Lucantoni

, et al. Leptin plasma concentrations are dependent on body fat distribution in obese patients. Int J Obes Relat Metab Disord, 2000; 24:1139–1144.

20.

Mullen

, Davenport

, Hutter

, et al. Impact of body mass index on perioperative outcomes in patients undergoing major intra-abdominal cancer surgery. Ann Surg Oncol, 2008; 15:2164–2172.

21.

Kerkhoffs

, Servien

, Dunn

, et al. The influence of obesity on the complication rate and outcome of total knee arthroplasty: A meta-analysis and systematic literature review. J Bone Joint Surg Am, 2012; 94:1839–1844.

22.

Namba

, Inacio

, Paxton

. Risk factors associated with surgical site infection in 30,491 primary total hip replacements. J Bone Joint Surg Br, 2012; 94:1330–1338.

23.

Wloch

, Wilson

, Lamagni

, et al. Risk factors for surgical site infection following caesarean section in England: Results from a multicentre cohort study. BJOG, 2012; 119:1324–1333.

24.

Kelesidis

, Kelesidis

, Mantzoros

. Adiponectin and cancer: A systematic review. Br J Cancer, 2006; 94: 1221–1225.