Length of Stay and Cost for Surgical Site Infection after Abdominal and Cardiac Surgery in Japanese Hospitals: Multi-Center Surveillance

Abstract

Purpose:

This study evaluated the influence of surgical site infections (SSIs) after abdominal or cardiac surgery on the post-operative duration of hospitalization and cost.

Methods:

A retrospective 1:1 matched case-control study of length of stay and healthcare expenditures for patients who were discharged from nine hospitals, between April 1, 2006 and March 31, 2008, after undergoing abdominal or cardiac surgery and who did and did not have a SSI.

Results:

Information was obtained from 246 pairs of patients who had undergone abdominal surgery and 27 pairs of patients who had undergone cardiac surgery. Overall, the mean post-operative hospitalization was 20.7 days longer and the mean post-operative healthcare expenditure was $8,791 higher in the SSI group than for the SSI-free group. Among the patients who had undergone abdominal surgery, development of SSI extended the average hospitalization by 17.6 days and increased the average healthcare expenditure by $6,624. Among the patients who had undergone cardiac surgery, SSI extended the post-operative hospitalization by an average of 48.9 days and increased the post-operative healthcare expenditure by an average of $28,534.

Conclusions:

Under the current healthcare system in Japan, the development of SSI after abdominal surgery necessitates extension of hospitalization two-fold and increases the post-operative healthcare expenditure 2.5-fold. Development of SSI after cardiac surgery necessitates extension of hospitalization fourfold and increases the healthcare expenditure six-fold.

Surgical site infection (SSI) is the third leading type of healthcare-associated infection, estimated to account for 14%–16% of all such infections [1]. Patients who develop SSIs are reported to be at a 2- to 11-fold higher risk of death, and SSIs are one of the clinically important post-operative complications [1,2]. Infection also is a serious post-operative complication from the viewpoint of medical economics. In the United States, on average, the development of SSIs was reported in 1980 to necessitate a 10-day extension of hospitalization and entail an additional expenditure of $2,000 per patient [3]. According to the reported for 1992, the development of SSIs extended the post-operative hospitalization by 7.3 days and increased the healthcare expenditure by $3,152 on average [4]. In 2009, the development of SSIs extended hospitalization by 9.7 days and increased the healthcare expenditure by $20,842 per patient on average, causing an additional inpatient care cost of $1.6 billion per year in total [5].

In Japan also, SSIs are recognized as one of the most important healthcare-associated infections. According to the information published by the Japan Nosocomial Infection Surveillance (JANIS) program, a program of the Ministry of Health, Labour and Welfare (MHLW), during the first half of 2009, SSIs developed in 2,743 of the 41,340 surgically treated patients surveyed, which represents an incidence rate of 6.6% [6].

Since the U.S. Centers for Disease Control and Prevention (CDC) published the Guidelines for Prevention of Surgical Site Infection in 1999 [7], increasing attention has been paid in Japan to the prevention of SSIs. Numerous reports have been published pertaining to methods for the prevention of SSIs and the incidence of SSIs. However, reports on the medical economic impact of SSIs have been confined to a few medical facilities and surgical procedures, and few reports of multicenter studies of this aspect of SSIs have been published. Meanwhile, the Diagnosis Procedure Combination/Per-Diem Payment System (DPC/PDPS) was introduced in Japan in 2003 for calculation of the amount to be reimbursed by insurers for acute inpatient care. Under such circumstances, there is a need for medical economic evaluation of the measures for the prevention of SSIs from the viewpoint of efficient utilization of available medical resources. Reports pertaining to the impact of SSIs on the medical economy from overseas are difficult to extrapolate to the circumstance prevailing in Japan, because of differences among countries in the healthcare systems, economy, foreign exchange rates, and so on. Thus, domestic data must be obtained.

The Japan Society for Surgical Infection undertook the present multi-center study to evaluate the impact of SSIs developing after common surgical procedures on the post-operative duration of hospitalization and post-operative healthcare expenditure. In this paper, we present briefly the findings of a comparison between patients undergoing abdominal surgery and those having cardiac surgery, as well as the results of an overall analysis of patients undergoing abdominal surgery.

Patients and Methods

Study design

This study was conducted with the approval of the ethics committee of each of the participating medical facilities. Information was obtained in a non-linkable anonymous form. The study was a multi-center retrospective matched case-control study involving 1:1 matching of cases (patients having SSI) and controls (patients who did not develop an SSI). The study period was from October 2009 to April 2010 (Table 1). The subjects were patients discharged from hospitals during the two-year period, from April 1, 2006 to March 31, 2008, after undergoing any of the operative procedures corresponding to 32 operation codes listed in the Medical Fee Reimbursement Schedule for 2006 (Table 2).

Table 1.

Participating Hospitals

Hospital	Region	Type	No. of beds	Procedure type
Teine Keijinkai Hospital	Hokkaido	Private	547	Abdominal
Sanin Rosai Hospital	Tottri	Public	383	Abdominal
Toho University, Medical Center Ohashi Hospital	Tokyo	Private	468	Abdominal
NTT Medical Center, Tokyo	Tokyo	Private	665	Abdominal
Yamaguchi University Hospital	Yamaguchi	Public	736	Abdominal
Toyonaka Municipal Hospital	Osaka	Public	613	Abdominal
National Cerebral and Cardiovascular Center Hospital	Osaka	Public	640	Cardiac
Kagoshima University Medical and Dental Hospital	Kagoshima	Public	725	Cardiac
Osaka University Hospital	Osaka	Public	1,076	Cardiac

Table 2.

Procedures Performed and Codes

	Procedure group	Code	Procedure
Heart surgery	CABG	K5521	Coronary artery, aortic bypass grafting (one anastomosis)
		K5522	Coronary artery, aortic bypass grafting (two or more anastomoses)
		K552-21	Coronary artery, aortic bypass grafting (without artificial heart–lung system) (one anastomosis)
		K552-22	Coronary artery, aortic bypass grafting (without artificial heart–lung system) (two or more anastomoses)
	Valve replacement/plasty	K5541	Valvuloplasty (one valve)
		K5542	Valvuloplasty (two valves)
		K5543	Valvuloplasty (three valves)
		K5551	Valve replacement (one valve)
		K5552	Valve replacement (two valves)
		K5553	Valve replacement (three valves)
Abdominal surgery	Total gastrectomy	K6572	Total gastrectomy (surgery for malignant tumor)
	Total gastrectomy (lap)	K657-22	Laparoscopic total gastrectomy (surgery for malignant tumor)
	Hepatectomy	K6951	Hepatectomy (partial resection)
		K6952	Hepatectomy (segmentectomy)
		K6953	Hepatectomy (lobectomy)
	Pancreatoduodenectomy	K7031	Pancreatic head tumor resection (pancreatoduodenectomy)
		K7032	Tumor resection with lymph node/nerve plexus excision, or duodenum-preserving pancreatic head resection
		K7033	Tumor resection with combined resection of surrounding organs (e.g., stomach, colon, kidney, adrenal)
		K7034	Tumor resection with revascularization
	Appendectomy	K718	Appendectomy
	Appendectomy (lap)	K718-2	Laparoscopic appendectomy
	Colectomy	K7193	Colectomy (total resection, subtotal resection or surgery for malignant tumor)
	Colectomy (lap)	K719-3	Laparoscopic resection of malignant colon tumor
	Proctectomy	K7402	Rectal resection/excision (low anterior resection)
	Proctectomy (lap)	K7403	Rectal resection/excision (ultralow anterior resection) (perianal coloanal anastomosis)
		K7404	Rectal resection/excision (excision)
		K740-2	Laparoscopic proctectomy
	Hysterectomy	K740-3	Laparoscopic low anterior resection of rectum
		K877	Total hysterectomy
	Cesarean section	K8981	Cesarean section (emergency)
		K8982	Cesarean section (elective)

CABG=coronary artery bypass grafting; lap=laparoscopy.

Case identification

Patients who developed SSIs post-operatively served as the cases. Cases were explored retrospectively using the patient record database at the medical facilities, as well as the SSI surveillance records. The diagnosis of SSI was made retrospectively by the attending physician or the reporter in accordance with the criteria for defining a SSI contained in the CDC Guidelines for Prevention of Surgical Site Infection. The presence/absence of SSIs was checked for 30 days post-operatively in patients who did not receive any implant and for one year post-operatively in patients who received an implant. Cases were seen between April 1, 2006 and March 31, 2008 and were identified by the reporter in order of the earlier day of operation.

Control matching

Case-control matching was carried out by the reporter for each medical facility. The criteria for matching were identical operation code, identical sex, age±5 years, presence/absence of preoperative corticosteroid use, and operation date closest to the case. However, for patients who had undergone coronary artery bypass grafting (CABG), matching according to the first three digits of the operation code was permitted. For patients who had undergone valvular surgery, matching between valve replacement surgery and valvuloplasty also was accepted. For these two cardiac operations, the presence/absence of identical accompanying operation and the presence/absence of pre-operative chronic dialysis sessions were adopted as additional criteria for matching. Matching of the same control with multiple non-SSI cases was avoided.

Information collection

Information on the post-operative length of hospitalization was collected from the patient record database at each participating facility. Hospitalization was defined by counting the number of days from the day after the operation (Day 1) to the day of discharge, referral, or death. For patients hospitalized again because of SSI, the number of days from readmission to discharge, referral, or death was added to the initially calculated length of hospitalization.

Post-operative healthcare expenditure was calculated by the fee-for-service method, covering the expenses for in-patient care during the hospitalization in accordance with the medical fee reimbursement schedule for 2006. Thus, the information collected did not cover the expenses for inpatient care before and on the day of the operation, operation fees, fees for anesthesiology services, or expenses for post-discharge outpatient care. The information on post-operative healthcare expenditure was collected for each category of healthcare used in the billing to the insurer, making use of the medical accounting database and receipt information maintained at each medical facility or the E file of the DPC survey data to be submitted to the MHLW (Table 3). Of the care categories, the categories with the code numbers “2x” (e.g., oral medication, as-needed medication, topical medication, dispensing, prescription, narcotic/toxic, basic dispensing fee) were combined into a single category of “prescription,” and the other categories were combined according to the second digit of the codes. The expenses for drugs and medical supplies consumed for each category were added to the expenses for the category. The post-operative expenses for additional operations not related to the SSIs and relevant expenses related to anesthesia were excluded. The US$/J¥ exchange rate adopted was $1=100 yen.

Table 3.

Treatment Category

Care category for billing	Name	Care category for present survey
21	Oral medication	Prescription
22	Bolus medication
23	Topical medication
24	Dispensing (inpatient)
25	Prescription
26	Narcotic/poison
27	Basic dispensing fee
28	Others (inpatient)
31	Subcutaneous/intramuscular	Injection
32	Intravenous
33	Others
39	Medication fee deducted
40	Manipulation	Manipulation
50	Surgery	Surgery
54	Anesthesia
60	Tests	Tests
70	Diagnostic imaging	Diagnostic imaging
80	Others	Others
90	Basic fee for inpatient care (inpatient)	Inpatient care
92	Specific fee for inpatient care (inpatient)

In all subjects, information was collected about the operation code, sex, age, pre-operative American Society of Anesthesiologists (ASA) score, post-operative length of hospitalization, and post-operative healthcare expenditure for each care category. For patients who developed SSI, additional information was collected from the patient record database as to the site, the bacteria isolated from the affected area, the presence/absence of reoperation for treatment of the SSI, and the presence/absence of readmission for management of the SSI.

Exclusion criteria

The study excluded patients who had developed infection at distant sites, ileus, respiratory failure, post-operative bleeding, or other serious complications not associated with SSIs. In addition, patients with ASA scores>3 points before abdominal surgery and patients who had undergone emergency heart surgery were excluded.

Statistical analysis

Statistical analysis was carried out with the Student t-test or Mann-Whitney U test using IBM SPSS Statistics 18 (IBM Japan, Tokyo, Japan).

Results

Information was obtained from nine medical facilities covering 546 patients (273 pairs). Information pertaining to cardiac surgery was collected from three facilities enrolling 54 patients (27 pairs), whereas information pertaining to abdominal surgery was collected from six facilities enrolling 246 pairs (Table 4).

Table 4.

Characteristics of Matched Groups

		SSI− (N=273 [%])	SSI+ (N=273 [%])	Total (N=546)
Male/female		184 (67)/89 (33)	184 (67)/89 (33)	368/178
Mean age at surgery (years)		63.4	63.5
ASA score	1	117 (43)	105 (38)	222
	2	139 (51)	150 (55)	289
	3	15 (5)	16 (6)	31
	4	2 (1)	2 (1)	4
Procedure	CABG	19 (7)	19 (7)	38
	Total gastrectomy	26 (10)	26 (10)	52
	Total gastrectomy (lap)	1 (<1)	1 (<1)	2
	Hepatectomy	22 (8)	22 (8)	44
	Colectomy	69 (25)	69 (25)	138
	Colectomy (lap)	23 (8)	23 (8)	46
	Hysterectomy	2 (1)	2 (1)	4
	Appendectomy	22 (8)	22 (8)	44
	Appendectomy (lap)	9 (3)	9 (3)	18
	Proctectomy (lap)	7 (3)	7 (3)	14
	Proctectomy	42 (15)	42 (15)	84
	Valve replacement/repair	8 (3)	8 (3)	16
	Pancreatoduodenectomy	23 (8)	23 (8)	46
Outcome	Death	0	3 (1)	3
	Discharge	270 (99)	261 (96)	531
	Change of hospital	3 (1)	9 (3)	12
Re-admission	+	0	9 (3)	9
	−	273 (100)	264 (97)	537
Second operation	+	0	41 (15)	41
	−	273 (100)	232 (85)	505
Steroid administration	−	260 (95)	263 (96)	523
	+	13 (5)	10 (4)	23

ASA=American Society of Anesthesiologists; lap=laparoscopy; CABG=coronary artery bypass grafting.

Overall, the mean post-operative hospitalization was 36.3 days in the patients who developed SSIs and 15.5 days in those who did not, a difference of 20.7 days (95% confidence interval [CI] 16.1, 25.3) (p<0.001). Among the patients who had undergone heart surgery, the mean hospitalization was 65.6 days in the SSI group and 17.1 days in the SSI-free group, a difference of 48.5 days (95% CI 39.6, 57.4) (p<0.001). Among the patients who had undergone abdominal surgery, the mean hospitalization was 33.1 days in the SSI group and 15.4 days in the SSI-free group, a difference of 17.7 days (95% CI 12.9, 22.5)(p<0.001). Overall, the mean post-operative healthcare expenditure was $13,237 in the SSI group and $4,446 in the SSI-free group, a difference of $8,791 (95% CI $6,440, $11,142)(p<0.001). Among the patients who had undergone cardiac surgery, the mean healthcare expenditure was $34,429 in the SSI group and $5,895 in the SSI-free group, a difference of $28,534 (95% CI $14,282, $42,786) (p<0.001). Among the patients who had undergone abdominal surgery, the mean post-operative healthcare expenditure was $10,911 in the SSI group and $4,287 in the SSI-free group, a difference of $6,624 (95% CI $4,682, $8,566)(p<0.001; Table 5).

Table 5.

Post-Operative Length of Stay and Cost (US$)

						95% CI for difference in mean
		SSI	Median	Mean	Difference in mean	Lower	Upper	p value^a
Total	LOS (days)	−	13	15.5	20.7	16.1	25.3	<0.001
		+	25	36.3
	Cost	−	3,564	4,446	8,791	6,440	11,142	<0.001
		+	7,057	13,237
Heart surgery	LOS (days)	−	16	17.1	48.5	39.6	57.6	<0.001
		+	62	65.6
	Cost	−	5,148	5,895	28,534	14,608	42,787	<0.001
		+	24,564	34,429
Abdominal surgery	LOS (days)	−	13	15.4	17.7	12.9	22.5	<0.001
		+	23	33.1
	Cost	−	3,362	4,287	6,624	4,682	8,566	<0.001
		+	6,518	10,911

Mann-Whitney U test.

CI=confidence interval; LOS=length of post-operative stay; SSI=surgical site infection.

Among the 246 pairs of patients who had undergone abdominal surgery, the mean healthcare expenditure for each care category was higher in the SSI group, an increase of $3,142 for inpatient fees, $1,590 for injection fees, and $718 for operation fees. When analyzed by the type of SSI, the mean post-operative length of hospitalization was 21.1 days for superficial incisional SSIs, 30.7 days for deep incisional SSIs, and 47.1 days for organ/space SSIs. The mean post-operative healthcare expenditure for patients with these three types of SSIs was $6,338, $9,497, and $16,374, respectively. When analyzed by the type of bacteria isolated, the mean post-operative length of hospitalization was 53 days for patients with infections caused by methicillin-resistant bacteria and 31 days for patients with infections caused by other bacteria, both being longer than the hospitalization for the SSI-free group. The mean post-operative healthcare expenditure was $18,707 for the patients with infections caused by methicillin-resistant bacteria and $10,221 for patients with infections caused by other bacteria, both being higher than the expenditure for the SSI-free cases (Table 6).

Table 6.

Mean Post-Operative Length of Stay (LOS) and Cost (US$) by Treatment Category in Abdominal Surgery

			SSI+
					Type			Type of bacteria isolated
		SSI−	Total	Difference	Superficial incisional	Deep incisional	Organ/space	Other	Methicillin-resistant
No.		246	246		125	13	108	226	20
Mean LOS		15.4	33.1	17.7	21.1	30.7	47.1	31.3	53.2
Mean costs		4,287	10,911	6,624	6,338	9,497	16,374	10,221	18,707
Treatment category	Prescription	129	383	254	185	343	618	$361	670
	Injection	569	2,159	1,590	908	1,898	3,638	1,733	4,956
	Treatment	76	450	374	192	182	781	352	924
	Surgery	68	786	718	277	405	1,421	628	1,457
	Examination	272	447	175	320	479	590	360	703
	Imaging	194	490	297	272	441	748	407	667
	Others	54	125	71	72	170	182	95	181
	Hospitalization	2,926	6,068	3,142	4,112	5,580	8,391	5,261	9,047

Discussion

In this retrospective study based on an existing database, we analyzed only the post-operative duration of hospitalization and the healthcare expenditure to avoid the confounding influence of the pre-operative length of hospitalization. Healthcare expenditure was calculated by the fee-for-service method, which represents the cost of the resources needed for healthcare services.

In this study, onset of SSI resulted on average in extension of the post-operative length of hospitalization by 20.7 days and an increase in the post-operative healthcare expenditure by $8,791. In the comparison of patients undergoing abdominal surgery with those undergoing heart surgery, the development of SSIs by the former was associated with an approximately two-fold increase in the length of hospitalization and a 2.5-fold increase in the healthcare expenditure, whereas that after heart surgery was associated with a 3.8-fold increase in the length of hospitalization and a 5.8-fold increase in healthcare expenditure. Thus, the development of SSIs after cardiac surgery imposed a greater burden on both healthcare insurance and the medical facilities.

Table 7 compiles the overseas reports of the burdens on the medical economy imposed by SSIs. Because of differences in the protocols, foreign currency exchange rates, economy, medical fee reimbursement system, timing of research, and other factors among countries, accurate medicoeconomic comparisons of hospital-associated infections at various sites are challenging [12]. However, consistent with our finding, the reports on multiple surgical procedures by Coello et al. [8] and de Lissovoy et al. [5] demonstrated a tendency for SSIs developing after heart surgery to impose a greater medical economic burden. This finding would seem to be attributable to the fact that the onset of SSIs such as mediastinitis after cardiac surgery is a serious clinical condition.

Table 7.

International Comparison of Burden of Surgical Site Infection (Literature Review)

				Mean length of stay				Median length of stay			Mean cost (US$)				Median cost
Study	Country	Year of study	Surgical procedure	SSI−	SSI+	Difference	% Increase	SSI−	SSI+	% Increase	SSI−	SSI+	Difference	% Increase	SSI −	SSI+	Increasing %
Coello et al. [8]	UK	1997–2001	Small bowel surgery	11.5	23.8	12.3	207						5,754
			Large bowel surgery	11.3	22	10.7	195						4,098
			CABG	7.4	21.7	14.3	293						5,841
Tanner et al. [9]	UK	2008	Colon surgery			14							15,785
de Lissovoy et al. [5]	US	2005	Colorectal	10.2	19.1	8.9	187	8	15	188	20,441	38,396	17,955	188	9,519	22,804	240%
			Gastrointestinal	7	17	10	243	4	12	300	16,239	37,068	20,829	228	14,172	27,794	196%
			Cardiovascular	10.2	23.9	13.7	234	8	19	238	41,066	78,579	37,513	191	31,652	60,891	192%
Kirkland et al. [2]	US	1991–1995	CABG					8	19						11,559	15,415
			Appendectomy					2	12						1,039	4,948
			Colon surgery					8	14						3,460	6,131
Eagye et al. [10]	US	2005–2007	Colorectal surgery (organ space infection)	6.2	21.7	15.5	350				10,999	42,516	31,517	387
Weber et al. [11]	Switzerland	2000–2001	Visceral surgery			17.2							14,992
Present series	Japan	2006–2008	Abdominal surgery	15.4	33.1	17.7	215	13	23	177	4,287	10,911	6,624	255	3,362	6,518	194%
			Cardiac surgery	17.1	65.6	48.5	384	16	62	388	5,895	34,429	28,534	584	5,148	24,564	477%

CABG=coronary artery bypass grafting; SSI=surgical site infection.

UK=United Kingdom; US=United States.

Our findings suggest a greater increase in the duration of hospitalization after the development of SSIs—both in patients undergoing abdominal surgery and in those undergoing cardiac surgery—than has been reported from other countries. According to the OECD Health Data, the mean duration of hospitalization for patients receiving acute care is longer in Japan than in any other developed country. It seems likely that the longer hospitalization of patients who develop SSIs is attributable to the tendency of Japanese patients to remain in the hospital until healing of an SSI is almost complete. Thus, the post-operative length of hospitalization increased by as many as 48.5 days in the patients who had SSIs after heart surgery. This seems to be attributable to the following factors: (1) SSIs following heart surgery (e.g., mediastinitis) often are difficult to treat; and (2) patients tend to continue treatment at acute care hospitals until complete healing of the infection even when the SSI is relatively mild (e.g., superficial incisional infection).

Patients who developed SSIs after abdominal surgery showed a hospitalization longer by 17.7 days on average. Of the additional post-operative healthcare expenditure incurred for these patients, 47% pertained to hospital overhead, 24% to parenteral medication, and 11% to operation. When analyzed by SSI type, the post-operative length of hospitalization and healthcare expenditure increased from superficial incisional infection to deep incisional infection to organ/space infection, as would be expected. The percentage of the total post-operative health care expenditure for non-parenteral and parenteral medication was 16% in the SSI-free group, but 17%, 23%, and 27%, respectively, in the patients who developed the aforementioned three types of SSI. Thus, the development of SSI had a greater influence on the expenses for medication than on the duration of hospitalization. This result suggests that SSIs in deeper tissues may impose a greater economic burden on medical facilities. A major factor responsible for the increase in the expenses for medication is the need for antimicrobial drugs. The present study did not collect information about the specific drugs used. At present, many medical facilities providing acute care in Japan are supplying detailed information about healthcare services (including information about the drugs used) to the MHLW. If such information were combined with the information about the diagnosis of SSIs, it would become possible to conduct a more detailed survey of post-operative healthcare expenditure.

It is well known that prevention of infection with drug-resistant pathogens is one of the most important measures against healthcare-associated infections [13]. The present study included patients from whom methicillin-resistant bacteria were isolated after abdominal surgery. Compared with the patients who did not develop SSIs after abdominal surgery, the post-operative hospitalization was 3.5-fold longer (53.2 days on average), and the healthcare expenditure was 4.4-fold higher ($18,707 on average), in the patients from whom methicillin-resistant bacteria were isolated. Also, compared with patients with SSIs from whom bacteria other than methicillin-resistant organisms were isolated, the post-operative hospitalization was 22 days longer and the expenditure in each category was higher in the patients with SSIs after abdominal surgery from whom methicillin-resistant bacteria were isolated. The expenses for non-parenteral medication in the latter patients were 8.7-fold higher than in the SSI-free group and 2.9-fold higher than in the SSI group from which other bacteria were isolated. These differences seem to be associated closely with the use of anti-MRSA drugs. Considering the high percentage of methicillin-resistant strains among staphylococci in Japan, these results illustrate the importance of measures against methicillin-resistant bacteria from the viewpoints of healthcare insurance finance and management of medical facilities, as well as patient outcomes.

On the basis of the results of this study, we estimated the influence of SSIs on medical resources throughout Japan. According to the report by Sakata et al. [14], the annual number of patients undergoing cardiac surgery of the types covered by this study is about 36,000. According to the JANIS report, the estimated rate of SSIs after heart surgery is 3%. We therefore estimate that medical resources equivalent to $31 million (52,380 beds×days/year) are utilized to deal with SSIs developing after heart surgery. According to the MHLW statistics from 2008 [15], the annual number of patients undergoing abdominal surgery of the types covered by this study is 346,000, and according to the JANIS report, the estimated incidence of SSIs following such abdominal surgery is 12%. We therefore estimate that medical resources equivalent to $275.5 million (736,000 beds×days/year) are utilized to deal with SSIs developing after abdominal surgery. Taken together, these data suggest that about $300 million in medical resources is utilized in Japan to deal with SSIs after surgery of the types covered by this study.

Measures to prevent peri-operative SSIs often are adopted by individual Japanese medical facilities based on the CDC/HICPAC Guidelines for the Prevention of Surgical Site Infections [7]. A reduction in the incidence of SSIs has been documented in many domestic reports. Reducing the incidence of SSIs and preventing the persistence of SSIs seem to contribute, to some degree, to improving the quality of acute-stage care and shortening hospitalization. The decrease in the use of injectable antibacterial drugs as a result of the shortened dosing period of prophylactic antimicrobial medications and the decrease in the use of therapeutic antimicrobial drugs as a result of reduction in the incidence of infections seem to have reduced healthcare expenditure under the health insurance system in Japan. It was difficult to evaluate these factors from a medical economic perspective.

In the past, few data were available in Japan about the medicoeconomic burden entailed in SSI prevention measures. We hope the present report will contribute to estimation of these costs.

This study, conducted retrospectively, has several limitations. First, no well-defined criteria for discharge from the hospital were available. This limitation, reflecting the status of healthcare in Japan during the period of the study, suggests that the situations faced by individual patients or medical facilities may have had some influence on the length of hospitalization. Second, this study excluded patients with distant metastasis, ileus, respiratory failure, post-operative bleeding, or other serious complications not associated with SSIs. However, it was not possible to eliminate completely the influence of post-operative complications rated as non-serious by the reporter. For this reason, we cannot say definitely that the data on the length of hospitalization and healthcare expenditure pertained exclusively to SSIs. Furthermore, this study involved a bias in patient selection, because the diagnosis of SSIs was made by the attending physician or the reporter on the basis of existing records. It is probable that the cases had severe SSIs, because records of cases with mild SSIs were absent. In addition, we cannot rule out biases attributable to the lack of non-randomized selection of the controls matched to the cases.

Third, this study pertained only to the increase in the post-operative length of hospitalization and to healthcare expenditure. It does not reflect the additional expenses for SSIs after discharge, treatment for SSIs at an outpatient clinic after discharge from the hospital, or treatment for SSIs after admission to other medical facilities. Note needs to be taken of this limitation as well.

Conclusion

The study revealed that under the current healthcare system in Japan, the development of SSIs after abdominal surgery is associated with a twofold increase in the duration of post-operative hospitalization and a 2.5-fold increase in post-operative healthcare expenditure and that the corresponding values after heart surgery are fourfold and sixfold, respectively. In addition, the medical economic burden posed by the development of SSIs after abdominal surgery increased as the site of the SSI became deeper, and the development of SSIs caused by methicillin-resistant bacteria was associated with a 3.6-fold increase in post-operative hospitalization and a 4.6-fold increase in the post-operative healthcare expenditure relative to the SSI-free group.

Footnotes

Acknowledgment

The authors are indebted to the participating researchers at the medical facilities, who provided information for this study, and to those in charge of information management at the medical information department, the medical administration section, and so on of the participating facilities for their cooperation.

Author Disclosure Statement

This study was carried out by the Clinical Trial Committee of the Japan Society for Surgical Infection under assignment from Johnson & Johnson K.K.

References

Engemann

, Carmeli

, Cosgrove

et al. Adverse clinical and economic outcomes attributable to methicillin resistance among patients with Staphylococcus aureus surgical site infection. Clin Infect Dis, 2003; 36:592–598.

Kirkland

, Briggs

, Trivette

et al. The impact of surgical-site infections in the 1990s: Attributable mortality, excess length of hospitalization, and extra costs. Infect Control Hosp Epidemiol, 1999; 20:725–730.

Cruse

. Wound infection surveillance. Rev Infect Dis, 1981; 4:734–737.

Martone

, Jarvis

, Culver

, Haley

. Incidence and nature of endemic and epidemic nosocomial infections. Bennett

, Brachman

. Hospital Infections, 3rd. Boston: Little, Brown and Co., 1992; 577–596.

de Lissovoy

, Fraeman

, Hutchins

et al. Surgical site infection: Incidence and impact on hospital utilization and treatment costs. Am J Infect Control, 2009; 37:387–397.

Japan Nosocomial Infection Surveillance. http://www.nih-janis.jp/index.asp

Mangram

. Horan TC, Pearson ML, et al. Guidelines for the prevention of surgical site infection. Infect Control Hosp Epidemiol, 1999; 20:247–278.

Coello

, Charlett

, Wilson

et al. Adverse impact of surgical site infections in English hospitals. J Hosp Infect, 2005; 60:93–103.

Tanner

, Khan

, Aplin

et al. Post-discharge surveillance to identify colorectal surgical site infection rates and related costs. J Hosp Infect, 2009; 72:243–250.

10.

Eagye

, Nicolau

. Deep and organ/space infections in patients undergoing elective colorectal surgery: Incidence and impact on hospital length of stay and costs. Am J Surg, 2009; 198:359–367.

11.

Weber

, Zwahlen

, Reck

et al. Economic burden of surgical site infection at a European university hospital. Infect Control Hosp Epidemiol, 2008; 29:623–629.

12.

Yasunaga

, Ide

, Imamura

, Ohe

. Accuracy of economic studies on surgical site infection. J Hosp Infect, 2007; 65:102–107.

13.

Anderson

, Kaye

, Chen

et al. Clinical and financial outcomes due to methicillin-resistant Staphylococcus aureus surgical site infection: A multi-center matched outcomes study. PLoS One, 2009; 4,12:e8305.

14.

Sakata

, Fujii

, Kuwano

. Thoracic and cardiovascular surgery in Japan during 2008: Annual report by the Japanese Association for Thoracic Surgery. Gen Thorac Cardiovasc Surg, 2010; 58:356–383.

15.

Ministry Health and Labor Welfare. [Survey of Medical Care Activities in Public Health Insurance] (Jpn) http://www.e-stat.go.jp/SG1/estat/List.do?lid=000001055081. 2010 Sept. 20.

16.

Cruse

, Foord

. The epidemiology of wound infection: A 10-year prospective study of 62,939 wounds. Surg Clin North Am, 1980; 60:27–40.