Positive impact of perioperative oral management on the risk of surgical site infections after abdominal surgery: Sixteen universities in Japan

Abstract

Surgical site infections (SSI) are associated with increased morbidity and mortality rates. This study aimed to investigate the ability of perioperative oral management (POM) to reduce the risk of SSI in abdominal surgery Real-world data collected from 16 university hospitals in Japan were reviewed. The medical records of consecutive 2782 patients (1750 men and 1032 women) who underwent abdominal surgery under general anesthesia at 16 university hospitals were retrospectively reviewed. Detailed information about SSI was assessed and compared between patients with and without POM in univariate and multivariate analyses. SSI were observed in 275 patients (incidence rate:9.9%), and POM was administered to 778 patients (28.0%). Univariate analyses revealed that diabetes mellitus, Eastern Cooperative Oncology Group performance status, American Society of Anesthesiologists classification, surgical site, preoperative Prognostic Nutritional Index score, POM, extent of surgery, operation time, and intraoperative blood loss were significantly associated with postoperative SSI (Chi-square or Mann–Whitney U test, P < .01). Multivariate analysis revealed that POM had significant preventive effects against postoperative SSI (estimate: −0.245, standard error: 0.080, P < .01). Surgical site, American Society of Anesthesiologists classification, and operation time were also significant and independent clinical predictors of SSI. The analysis of real-world data from 16 university hospitals revealed that, regardless of the content and degree of the problem, the addition of POM has significant beneficial effects in reducing the risk of SSI in patients who undergo abdominal surgery. Medical records from each hospital and data from the Health Care Payment Fund were collected and analyzed retrospectively.

1. Introduction

Surgical site infections (SSI) are associated with increased morbidity and mortality.^[1] SSI influences the duration of hospitalization and healthcare expenditure and has a negative impact on patients’ quality of life. In a report on nosocomial infection surveillance in Japan, the incidence of SSI was estimated to be approximately 6%.^[2] The report revealed that SSI occurred most frequently after abdominal surgery (digestive organ surgery), accounting for 88% of all SSI. SSI is one of the most common infectious complications after abdominal surgery.^[3] Thus, measures to prevent SSI are important.

A Surgeon General’s report on oral health in America showed that oral functional management can prevent general complications during the perioperative period.^[4] In Japan, the National Health Insurance System covers perioperative oral management (POM) after 2012, and patients scheduled to undergo surgery for cancer, cardiovascular disease, or organ transplantation receive dental and oral functional management during the perioperative period. POM might help prevent dental-related local and systemic infections in patients who are compromised by invasive surgery. POM has also been reported to prevent postoperative pneumonia, which has been attributed to aspiration of oropharyngeal pathogens.^[4] Recently, Nobuhara et al^[5] reported that POM reduced the risk of SSI after colorectal cancer surgery and shortened the postoperative hospital stay. The Centers for Disease Control and Prevention (CDC) guidelines recommend that all infections that arise at locations remote from planned surgical sites should be identified and treated before elective operations, and elective operations should be postponed in cases involving patients with remote site infections until the infection resolves.^[6] There might be a possibility that POM might help to reduce the frequency of SSI after abdominal surgery. However, no large studies on the effects of POM for abdominal surgery, including cancer and non-cancer patients, on the risk of SSI have been performed.

Therefore, the purpose of the present study was to investigate the ability of POM to prevent SSI in patients undergoing abdominal surgery under general anesthesia. Real-world data of consecutive patients who underwent abdominal surgery in 16 Japanese university hospitals were reviewed.

2. Materials and Methods

The present study was conducted by the Japanese Stomatological Society. The medical records of 2782 patients scheduled to undergo abdominal surgery under general anesthesia between April 2016 and March 2017 at 16 university hospitals were retrospectively reviewed, and data concerning the presence or absence of SSI and clinical factors that might influence the risk of SSI were collected. The presence/absence of SSI was determined based on documentation of the patients’ medical records. The potential predictors of SSI included patient factors (age, sex, smoking habits, diabetes mellitus, severe heart disease, severe pulmonary disease, Eastern Cooperative Oncology Group performance status [ECOG-PS], preoperative Prognostic Nutritional Index [PNI], treatment factors [risk of general anesthesia, surgical site, extent of surgery, operation time, and amount of intraoperative blood loss], and the presence or absence of POM). Severe heart disease was defined as ≥ grade 3 according to the New York Heart Association Functional Classification.^[7] Severe pulmonary disease was defined as a percentage vital capacity of <60% or a percentage forced expiratory volume in one second of <50%. The risk of general anesthesia was evaluated using the American Society of Anesthesiologists (ASA) physical status classification.^[8] The extent of surgery was divided into three groups: endoscopic surgery or an incision ≤10 cm, laparotomic surgery or an incision >10 cm, or laparotomy and thoracotomy. The surgical sites were divided into 10 groups: stomach/duodenum, jejunum/ileum/appendix/colon, rectum, liver, gallbladder, biliary tract, pancreas, abdominal wall/abdominal hernias, peritoneum/retroperitoneum/mesentery/retina, anus/periphery, and the spleen. PNI was calculated according to the method reported by Onodera et al^[9] PNI scores were classified into 50-point groups. Operation times were classified into 3-hour groups, and intraoperative blood loss was categorized into 300-mL groups.

POM was generally initiated when the decision to hospitalize the patient was made. POM included oral health instructions, removal of dental calculi (scaling), professional mechanical tooth cleaning, removal of tongue coatings, cleaning and adjustment of dentures, and extraction of teeth affected by severe periodontitis that exhibited pain, pus discharge, mobility, or marked alveolar bone loss on radiographic examination. POM could not be performed if the patient did not wish or if time constraints prevented the treatment.

The effects of POM on SSI were investigated statistically using univariate (chi-square test, Mann–Whitney U test) and multivariate (stepwise logistic regression analysis) analyses. Statistical analyses were performed using JMP ver.13 (SAS Institute Inc., Cary, NC). Statistical significance was set at P < .05.

3. Results

The results of stepwise multivariate analyses are summarized in Table 1. The surgical site, ASA classification, operation time, and POM induction had independent and significant effects on the manifestation of SSI. The effects of smoking habits on SSI were significant (P = .06). The obtained results revealed that POM helped prevent SSI (estimate, −0.245; standard error, 0.080; P < .01; odds ratio [OR], 0.78). On the other hand, other variables (surgery involving the pancreas, gallbladder, biliary tract, or liver vs others, OR: 1.30; ASA classification, OR: 1.38/class; and operation time, OR: 1.60/3 h) were significantly negatively associated with SSI.

Table 1.

A multivariate analysis of the effects of POMs on prevalence of surgical site infection in all subjects.

	Estimate	Standard error	Odds ratio	95% CI		Chi-square	P value
Smoking habit (continuing/stop smoking or never)	0.234	0.123	1.26	0.98	1.59	3.63	.06
Surgical site (Pancreas, gallbladder and biliary trac, and liver/others)	0.265	0.073	1.30	1.13	1.50	13.32	<.01
ASA classification*	0.319	0.118	1.38	1.09	1.73	7.31	<.01
Operation time (every 3 h)	0.473	0.070	1.60	1.40	1.84	46.11	<.01
Intervention of POMs (Yes/No)	−0.245	0.080	0.78	0.67	0.91	9.44	<.01

POM = perioperative oral management.

^*The American Society of Anaesthesiologists physical status classification

Patient characteristics are summarized in Table 2. The study included 1750 men and 1032 women. Patients aged 40 to 79 years accounted for 73.4% of the study population. As for ECOG-PS, 1694 (61.0%), 535 (19.2%), 154 (5.5%), 51 (1.8%), and 47 (1.7%) patients had ECOG-PS scores of 0, 1, 2, 3, and 4, respectively. Regarding ASA physical status classification, 462 (16.7%), 1781 (64.0%), 316 (11.4%), and 15 (5.3%) patients were categorized as Class 1, Class 2, Class 3, and Class 4, respectively. POM was performed in 778 patients (28.0%).

Table 2.

The characteristics of patients and results of univariate analysis.

Variables	No. of patients	Prevalence of SSI (%)	Result of univariate analysis	Variables	No. of patients	Prevalence of SSI (%)	Result of univariate analysis
Gender				Surgical site
Female	1032	9.5%	NS (chi-square test)	Stomach and duodenum	645	6.6%	P < .01 (chi-square test)
Male	1750	10.1%		Jejunum, ileum, appendix, and colon	563	8.2%
Age				Rectum	380	7.9%
0–9	152	1.6%	NS (Mann–Whitney U test)	Liver	376	13.2%
10s	27	7.4%		Gallbladder and biliary tract	234	14.0%
20s	39	15.4%		Pancreas	330	19.0%
30s	83	4.8%		Abdominal wall and hernia	136	6.1%
40s	202	11.4%		Peritoneum, retroperitoneum, mesentery, and retina	82	4.1%
50s	332	13.0%		Anus and periphery	13	8.3%
60s	790	9.1%		Spleen	23	8.7%
70s	717	10.6%		Preoperative PNI score
80<	259	10.0%		<50	27	23.1%	P < .01 (Mann–Whitney U test)
Diabetes mellitus				50≤ to <100	722	11.0%
No	2261	9.0%	P < .01 (chi-square test)	100≤ to <150	1052	10.0%
Yes	521	13.2%		150≤ to <200	261	7.3%
Smoking				200≤ to <250	68	2.9%
Never	1636	12.9%	NS (chi-square test)	250≤ to <300	38	5.3%
Stop smoking	794	9.3%		300≤	103	2.9%
Continuing	352	10.3%		Unknown	511	12.26
Performance status (ECOG)				Extent of surgery
Score 0	1694	9.3%	P < .01 (Mann–Whitney U test)	Endoscopic or incision ≤ 10 cm	1214	7.3%	P < .01 (Mann–Whitney U test)
Score 1	535	11.0%		laparotomy or incision > 10 cm	1212	12.5%
Score 2	154	15.6%		Laparostomy and thoracotomy	101	4.4%
Score 3	51	15.7%		Unknown	255
Score 4	47	4.30%		Operation time
Unknown	301			<3 h	306	8.2%	P < .01 (Mann–Whitney U test)
Sever heart disease				3≤ to <6 h	1574	6.3%
No (NYHA < 3)	2494	5.1%	NS (chi-square test)	6≤ to <9 h	629	13.5%
Yes (NYHA ≥ 3)	78	10.0%		9≤ to <12 h	214	21.8%
Unknown	210			12≤	59	33.9%
Sever pulmonary disease*				Blood loss
No	2061	9.8%	NS (chi-square test)	0–299 mL	1807	6.6%	P < .01 (Mann–Whitney U test)
Yes	86	9.0%		300–599	328	12.2%
Unknown	635			600–899	172	15.7%
ASA physical status classification				900–1199	83	24.1%
Class 1	462	7.1%	P < .01 (Mann–Whitney U test)	1200–1499	56	21.4%
Class 2	1781	9.8%		1500–1799	27	33.3%
Class 3	316	12.7%		1800–2099	29	17.2%
Class 4	15	33.3%		2100–2399	15	20.0%
Unknown	208			2400–2699	10	30.0%
				2700–2999	9	11.1%
				3000=<	38	36.8%
				Intervention of perioperative oral management
				No	2004	10.7%	P < .01 (chi-square test)
				Yes	778	8.0%

%FEV1 = percentage forced expiratory volume in one second, ASA = American Society of Anaesthesiologists, ECOG = Eastern Cooperative Oncology Group, NYHA = New York Heart Association, PNI = preoperative Prognostic Nutritional Index, VC = percentage vital capacity.

^*Severe pulmonary disease: No (%VC ≥ 60% or FEV1.0% ≥ 50%) and Yes (%VC < 60% or FEV1.0% < 50%).

SSI were observed in 275 patients (9.9%). The pancreas was the most common site of SSI (incidence: 19.0%), followed by the gallbladder, biliary tract (14.0%), and liver (13.2%) (Table 2). The results of the univariate analyses are summarized in Table 1. SSI was significantly associated with the presence of diabetes mellitus, ECOG-PS, ASA physical status, surgical site, preoperative PNI score, extent of surgery, operation time, and amount of intraoperative blood loss. Higher SSI prevalence rates were associated with diabetes mellitus, higher ECOG-PS scores, higher ASA physical status classes, particular surgical sites, lower preoperative PNI scores, absence of POM, more invasive surgery, longer operation times, and higher amounts of intraoperative blood loss. Sex, age, severe heart disease, and severe pulmonary disease were not significantly associated with the prevalence of SSI.

4. Discussion

Postoperative SSI is an important marker of surgical quality and is associated with longer hospital stay, higher rates of reoperation and readmission, and increased mortality. Poor survival was observed in patients with various types of cancer who exhibited SSI after surgery.^[10] It has also been suggested that SSI is associated with shorter disease-free and overall survival in several types of cancer, such as head and neck, breast, colon, and gastric cancers.^[2] Furthermore, Nespoli et al^[2,10] showed that postoperative infections are associated with a poor 5-year survival rate among patients with colon cancer. SSI is one of the most common postoperative morbidities, occurring in 5% to 40% of patients who undergo colorectal surgery. In addition, SSI occurs in 7.6% of patients after major abdominal surgery.^[11] In the present study, SSI occurred in 9.9% of patients, which is consistent with the findings of previous studies.^[11] As SSI occurs frequently, measures aimed at preventing it are required.

Many investigators have studied the factors that influence the prevalence of SSI after abdominal surgery. Togo et al^[12] revealed that SSI was significantly associated with a long operation time, marked intraoperative blood loss, high blood transfusion requirement, and bile fistulas in patients with liver cancer. Du et al^[13] demonstrated that SSI was significantly associated with the type of surgery, surgical procedure (laparoscopic or open), and hospital size (i.e., number of beds) in patients with colorectal cancer. Fukuda et al^[14] showed that SSI was significantly associated with blood transfusions, use of antidiabetic drugs, use of steroids, operation time, American Society of Anesthesiologists classification, surgical procedure, emergency surgery, and age. In the current study, we retrospectively reviewed the cases of patients who underwent abdominal surgery and found that hepatic, biliary, or pancreatic surgery, poorer ASA physical status, and longer operation time were significantly associated with a higher prevalence of SSI.^[12] Most risk factors are difficult to control preoperatively.

In contrast, this study revealed that POM interventions had independent, significant, and beneficial effects on the risk of SSI after abdominal surgery. Recently, Nobuhara et al reported that POM reduces the risk of SSI after surgery for colorectal cancer.^[4] Our findings were consistent with those of the latter study. Other studies have found that preoperative oral care reduces the risk of postoperative pneumonia after surgery for esophageal, lung, or major cancer.^[12,15,16] As aspiration of oral/oropharyngeal fluid containing pathogenic microorganisms is considered to be the main cause of postoperative pneumonia, it is logical that controlling oral bacteria reduces the risk of postoperative pneumonia. Similarly, our results suggest that POM may significantly reduce the risk of SSI after abdominal surgery.

Gingivitis and periodontal diseases provide opportunities for bacterial overgrowth, and the richly vascularized and often ulcerated tissues associated with these diseases are susceptible to bacterial invasion.^[17] The bulk of dental plaque, such as biofilms, is composed of microcolonies of oral bacteria.^[12] Nobuhara et al^[5] reported that the oral cavity is a significant reservoir of pathogenic microorganisms that can infect multiple organs. Oral bacteria are known to influence various diseases, such as pneumonia, cardiovascular disease, cerebrovascular disease, rheumatoid arthritis, preterm or low-weight births, sepsis, carcinogenesis, nonalcoholic steatohepatitis, and SSI.^[18,19] Therefore, achieving quantitative and qualitative control of oral bacteria via oral healthcare is important for preventing infectious diseases.

The direct transfer of oral bacteria might cause SSI after head and neck or upper digestive tract surgery as well as postoperative aspiration pneumonia.^[5] In addition, the intravascular invasion of odontogenic bacteria, inflammatory cytokines, and/or endotoxins and their transport to remote organs through blood vessels or lymph ducts might cause SSI at various surgical sites.^[5] It is well known that transient bacteremia often occurs after tooth extraction and tooth brushing. CDC guidelines suggest that preoperative infectious lesions at remote sites are a risk factor for SSI.^[6] In 2012, POM began to be covered by the Japanese national health insurance system and has since been widely performed in patients who are scheduled to undergo cancer treatment, organ transplantation, cardiovascular surgery, or orthopedic implant surgery. We assessed changes in the prevalence of oral bacteria detected in blood cultures before and after the introduction of insurance coverage for POM and found that the introduction of insurance coverage for POM had a beneficial effect on the frequency of systemic infections caused by oral bacteria.^[20]

In addition, we recently found that POM had significant positive effects on perioperative serum albumin levels in patients treated surgically under general anesthesia (submitted for publication). Decreased serum albumin level is an independent risk factor for severe postoperative complications and poor prognosis.^[21] Some researchers have reported that dental infections and poor oral health are closely associated with lower serum albumin levels. POM, including oral care, removal of chronic dental infections, and prosthodontic treatments, has positive effects on serum albumin levels, which might consequently reduce the risk of SSI after abdominal surgery.^[21]

One of the major advantages of the present study is that it is the first to clarify the beneficial effects of POM on the risk of SSI in patients who underwent abdominal surgery, based on a multicenter retrospective study involving a large number of cases. The limitation of the present study is the differences observed in POMs criteria and treatment protocols at each institute because of its retrospective nature. This is a retrospective, observational study, and the presence or absence of SSI is determined based on physician judgment; the determination of SSI generally follows CDC criteria, but because it is a retrospective study, the criteria are not rigorously standardized. In addition, the POM intervention rate was relatively low (28.0%). However, there are ethical difficulties associated with conducting a prospective randomized control study to evaluate the efficacy of POM because it has only been covered by the Japanese national health insurance system since 2012, and most Japanese patients now receive POM before undergoing cancer, cardiovascular, or transplant treatments. Furthermore, the protocols and aims of dental interventions are still subject to debate. Establishing guidelines for POM in patients scheduled to undergo surgery is needed to standardize such dental interventions.

5. Conclusion

The analysis of real-world data from 16 university hospitals revealed that regardless of the content and degree of the problem, the addition of POM has significant beneficial effects in reducing the risk of SSI in patients who undergo abdominal surgery. The protocols and aims of dental interventions are still subject to debate. Establishing guidelines for POM in patients scheduled to undergo surgery is needed to standardize such dental interventions.

Author contributions

Conceptualization: Hiroshi Kurita, Hideki Tanzawa.

Data curation: Tetsu Shimane, Kazuyuki Koike, Shigeyuki Fujita, Emiko Tanaka Isomura, Daichi Chikazu, Naomi Kanno, Keiichi Sasaki, Satoshi Hino, Hideharu Hibi, Takahiro Koyama, Seiji Nakamura, Takeshi Nomura, Yoshiyuki Mori, Itaru Tojyo, Toshiro Yamamoto, Iku Yamamori, Keiko Aota, Hideki Tanzawa.

Formal analysis: Hiroshi Kurita, Itaru Tojyo, Hideki Tanzawa.

Methodology: Hiroshi Kurita, Hideki Tanzawa.

Project administration: Hideki Tanzawa.

Resources: Hiroshi Kurita.

Software: Hiroshi Kurita.

Supervision: Hiroshi Kurita, Hideki Tanzawa.

Writing – original draft: Tetsu Shimane, Hiroshi Kurita.

Writing – review & editing: Hiroshi Kurita, Hideki Tanzawa.