Impact of Intra-Operative Adverse Events on the Risk of Surgical Site Infection in Abdominal Surgery

Abstract

Background:

Intra-operative adverse events (iAEs) recently were shown to correlate independently with an increased risk of post-operative death, morbidity, re-admissions, and length of hospital stay. We sought to understand further the impact of iAEs on surgical site infections (SSIs) in abdominal surgical procedures and delineate which patient populations are most affected. We hypothesized that all patients with iAEs have an increased risk for SSI, especially those with pre-existing risk factors for SSI.

Patients and Methods:

To identify iAEs, a well-described three-step methodology was used: (1) the 2007–2012 American College of Surgeons-National Surgical Quality Improvement Program database was merged with the administrative database of our tertiary academic center, (2) the merged database was screened for iAEs in abdominal surgical procedures using the International Classification of Diseases, Ninth Revision, Clinical Modification-based Patient Safety Indicator “Accidental Puncture/Laceration,” and (3) each flagged record was systematically reviewed to confirm iAE occurrence. Uni-variable and backward stepwise multi-variable analyses (adjusting for demographics, co-morbidities, type and complexity of operation) were performed to study the independent correlation between iAEs and SSIs (superficial, deep incisional, and organ-space). The correlation between iAEs and SSIs was investigated especially in patients deemed a priori at high risk for SSIs, specifically those older than age 60 and those with diabetes mellitus, obesity, cigarette smoking, steroid use, or American Society of Anesthesiologists class ≥III.

Results:

A total of 9,288 operations were included, and iAEs were detected in 183 (2.0%). Most iAEs consisted of bowel (44%) or vessel (29%) injuries and were addressed intra-operatively (92%). SSI occurred in 686 (7.4%) cases and included 331 (3.5%) superficial, 32 (0.34%) deep incisional, and 333 (3.6%) organ/space infections. iAEs were correlated independently with SSI (odds ratio [OR] = 1.67; 95% confidence interval [CI], 1.11–2.52, p = 0.013), and more severe iAEs were associated with a higher risk of infection. Analysis by SSI type revealed a significant association with organ/space SSI (OR = 1.81, 95% CI 1.07–3.05; p = 0.027), but not incisional infections. Most interestingly, the occurrence of an iAE was correlated with increased SSI rate in the low-risk but not the high-risk patient populations. Specifically, iAEs increased SSI in patients younger than 60 (OR = 2.69, 95% CI 1.55–4.67, p < 0.001), non-diabetic patients (OR = 1.64, 95% CI 1.04–2.58, p = 0.034), non-obese patients (OR = 2.9, 95% CI 1.81–4.66, p < 0.001), non-smokers (OR = 1.67, 95% CI 1.08–2.6, p = 0.022), with no steroid use (OR = 1.73, 95% CI 1.15–2.6, p < 0.008), and with ASA class <III (OR = 2.26, 95% CI 1.31–3.87, p = 0.003).

Conclusions:

The iAEs are associated independently with increased SSIs, particularly in patients with less pre-existing risk factors for SSI. Preventing iAEs or mitigating their impact, once they occur, may help decrease the rate of SSIs.

The annual incidence of surgical site infections (SSI) in patients undergoing inpatient surgical procedures in the United States is estimated to be 2% to 5% [1]. These estimates, however, likely under-report the true incidence of SSI and are highly variable, depending on procedure complexity [2 –4]. SSI negatively impact post-operative deaths and morbidity, and studies have demonstrated significant increases in the utilization of intensive care units, inpatient length of stay, and hospital re-admissions [5 –8]. In addition to the clinical burden of SSIs, the financial ramifications of SSIs are equally significant, with the World Health Organization estimating that they cost the United States up to $10 billion annually [9 –12]. Given the high volume of operations performed each year in the United States, efforts to decrease the occurrence of SSIs are critical [13].

Extensive research has defined patient-related risk factors for the development of post-operative SSI [14 –17]. Further, studies demonstrating intra-operative risk factors have led largely to the development of the Surgical Care Improvement Project (SCIP) quality measures to reduce the occurrence of SSI [18]. The association of intra-operative adverse events (iAEs), however, defined as inadvertent injuries that occur during the course of an operation, is not considered in current prediction models. iAEs have been shown to correlate independently with an increased risk of post-operative death, morbidity, re-admissions, length of hospital stay. and healthcare costs [19 –22]. While these studies suggest that patients with iAEs are at an increased risk of SSI, further characterization is needed. Identification of factors that increase the risk of SSI is essential, because stratification of patients into “high” and “low” risk groups can alter management and inform SSI prevention efforts.

Within this context, our study was conducted to provide a detailed account of the relationship between iAEs and SSI in patients undergoing abdominal surgical procedures. We sought to delineate how iAE severity impacts SSI and to determine which patient populations are most affected by the occurrence of an iAE. We hypothesized that all patients with iAEs have an increased risk for SSI—in particular, those with pre-existing risk factors for SSI.

Patients and Methods

To study the independent impact of iAEs on SSIs and delineate which patient populations are at the greatest risk, a well described four-step methodology was utilized: (1) Our 2007–2012 institutional American College of Surgeons-National Surgical Quality Improvement Program (ACS-NSQIP) and administrative databases for abdominal operations were matched retrospectively, and patient demographic and post-operative SSI data were extracted; (2) the matched database was screened for iAEs using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM)-based Patient Safety Indicator “Accidental Puncture/Laceration”; (3) a systematic review of identified charts was performed to confirm the occurrence of iAEs, classify their severity, and describe the complexity of the operation; and finally, (4) statistical analyses with uni-variable then logistic multi-variable regression models were constructed to assess the independent impact of iAEs on SSIs and to determine which patient populations are at the greatest risk. This study was approved by the Partners Institutional Review Board (2014P001456).

Patient population

All adult patients undergoing abdominal surgical procedures under general anesthesia in a tertiary care academic center from January 2007 to October 2012 were included. The hospital-wide comprehensive administrative database was linked with our institutional ACS-NSQIP database, and cases captured by both databases were selected for additional analysis. The ACS-NSQIP is a prospectively collected outcomes registry with a well-validated methodology, where an independent nurse reviewer systematically collects pre-specified pre-operative, intra-operative (not including iAEs), and post-operative variables. Before the start of the study, six pre-operative variables that are well-described risk factors for SSI were identified to determine “low-risk” and “high-risk” patient cohorts. These variables included age >60, diabetes mellitus, obesity, current cigarette smoking, chronic steroid use, and American society of Anesthesiologists (ASA) class ≥III.

Definition and classification of SSIs

The ACS-NSQIP database captures SSIs as a post-operative event. SSI is defined as an infection that occurs within 30 days of an operation [23]. SSIs require at least one of the following to be diagnosed: (1) Purulent drainage from the superficial incision; (2) organisms isolated from an aseptically obtained culture of fluid or tissue from the incision; (3) one symptom or sign of infection (pain or tenderness, localized swelling, redness, warmth, or fever); or (4) diagnosis made by the surgeon or attending physician. SSIs are further classified as superficial (i.e., skin or subcutaneous tissue), deep incisional (i.e., fascia and muscle), or organ/space (organ or body cavity that was manipulated during the operation). The SSI categories used by ACS-NSQIP align with those defined by the Centers for Disease Control and Prevention [24]. In our study, if more than one type of SSI developed in a patient, the infection was classified only according to the most severe infection.

Identification and classification of iAEs

An adverse event is defined as “an injury caused by medical management rather than the underlying disease.” We defined an iAE as an inadvertent injury during the operation. The matched database was queried using the ICD-9-CM-based algorithm for “accidental puncture or laceration” (APL) to identify occurrences of accidental intra-operative injury. The APL is the Agency for Healthcare Research and Quality's 15th Patient Safety Indicator, also known as PSI #15. The PSI #15 has been reported to have a positive predictive value ranging between 85% and 91% for procedure-related complications but is not specific to iAEs [20,22,25].

The operative notes for all PSI #15-identified cases were then reviewed independently by two investigators to confirm the occurrence of at least one iAE. The severity of the iAE (Class I–VI) was categorized using a previously validated severity classification system [20]. In this classification system, a “minor” iAE (Class I, II) includes injuries that required no repair within the same procedure (Class I), or surgical repair without either organ removal or a change in the originally planned procedure (Class II). A “major” iAE (Class III–VI) includes injuries for which repair necessitated tissue/organ resection with completion of the originally planned procedure (Class III), significant change in or incompletion of the originally planned procedure (excluding conversion from minimally invasive to open) (Class IV), re-operation within seven days of the index procedure (Class V), or intra-operative death because of hemorrhage (Class VI).

Intra-operative variables

Intra-operative variables were extracted from both the operative notes and the administrative database, including the type of surgical procedure performed, the phase of operation during which the iAE occurred (access, dissection, retraction, or resection and reconstruction), the surgical approach (open vs. laparoscopic), as well as the nature and severity of the iAE. The surgical approach was based on the intention to treat principle; laparoscopic operations that were converted to open were classified as “laparoscopic.” Adhesiolysis was assessed by querying the ACS-NSQIP database for procedure-specific Current Procedural Terminology (CPT) codes (44005, 44180, 50715, 58660). Operative complexity was assessed using, as a proxy, each procedure's relative value unit ([RVU]; by the Centers for Medicare and Medicaid Services Resource Based Relative ValueScale) based on CPT codes. The RVUs have been shown to be a better predictor of surgical outcomes than complexity scores established by panels [26,27]. They have also been shown in the ACS-NSQIP to predict independently post-operative morbidity in general surgical procedures [28 –30].

Statistical analysis

Unadjusted analyses were performed for pre-operative patient characteristics, intra-operative variables, and iAE details between patients with no SSI and SSI using chi-square test for categoric data and the Student t test for continuous data. Unless otherwise noted, results are reported as absolute number and percent (%) frequency.

Logistic multi-variable regression models were constructed to assess the independent impact of both any iAE and major iAE on SSI for patient populations deemed a priori at high risk for SSI compared with their lower risk counterparts. Specifically, a subset uni-variable analysis was performed for patients with each specific risk factor for SSI (i.e., diabetes mellitus). Variables potentially associated (p < 0.20) with SSI in the uni-variable analysis were included in the initial multi-variable models; variable selection for the final models was performed in a backward stepwise fashion. Laparoscopic approach, wound classification and adjusted RVU were included in each model, because these have been shown to be predictors of post-operative morbidity and wound infection. A multi-variable model was constructed for both the a priori higher risk patients (i.e., those with diabetes mellitus) and their potentially lower risk counterparts (i.e., those without diabetes mellitus). Statistical significance was set at p < 0.05. Statistical analysis was conducted using Intercooled Stata software, version 13.1 (StataCorp, College Station, TX). Statistical significance was accepted at the p value <0.05 level. Graphs were constructed using Prism version 7.0a.

Results

Patient population

A flowchart of patients included in this study, classified by iAE and SSI, is displayed in Figure 1. The total number of patients was 9,288, of whom 183 (2.0%) had a confirmed iAE. The rate of SSI in the patients who had an iAE was 18.6% (n = 34). This included 12 superficial SSIs (6.6%), three deep incisional SSIs (1.6%), and 19 organ/space SSIs (10.4%). Of the 9,105 patients who did not have an iAE, the rate of SSI was only 7.3% (n = 662). This included 319 superficial SSIs (3.5%), 29 deep incisional SSIs (0.3%), and 314 organ space SSIs (3.5%).

FIG. 1.

Flow diagram of study population divided by intra-operative adverse events (iAEs) and surgical site infection (SSI).

Pre-operative and intra-operative characteristics

Pre-operative and intra-operative characteristics of the entire patient population are displayed in the first column of Tables 1 and 2, respectively. Of the 9,288 patients included in the study, the mean age was 56 years, 54% were female, 89% were white, the mean body mass index was 31, and 35% had an ASA classification of III or greater. The most prevalent co-morbidities were hypertension (42%) and obesity (38%). The most common primary operation was intestinal (45%), followed by hepatopancreaticobiliary (28%). A minimally invasive operation was performed in 47% of cases, and 14% were emergent in nature. The median duration of the operations was 2.2 hours (interquartile range: 1.3–3.4), and adhesiolysis was performed in 18% of cases.

Table 1.

Uni-Variable Analyses of Pre-Operative Characteristics and Surgical Site Infection

	n (%) [Total n = 9,288]	No SSI [n = 8,592]	SSI [n = 696]	p
A priori risk factors
Age >60 y	3760 (40.5)	3392 (39.5)	368 (52.9)	<0.001
Diabetes mellitus	1431 (15.4)	1305 (15.2)	126 (18.1)	0.04
Obesity	3470 (38.2)	3219 (38.3)	251 (37.0)	0.503
Smoker	1411 (15.2)	1297 (15.1)	114 (16.4)	0.364
Long-term steroid use	385 (4.2)	342 (4.0)	43 (6.2)	0.005
ASA class ≥III	3288 (35.4)	2946 (34.4)	332 (47.7)	<0.001
Additional pre-operative characteristics
Age, years; mean (SD)	55.6 (16.8)	55.14 (16.8)	60.7 (15.1)	<0.001
Female	5043 (54.3)	4693 (54.6)	350 (50.3)	0.027
Body mass index; mean (SD)	30.5 (9.7)	30.6 (9.9)	29.4 (8.1)	0.003
Race
White	8285 (89.2)	7646 (89.0)	639 (91.2)	0.021
African American	328 (3.5)	306 (3.6)	22 (3.2)	0.582
Asian	232 (2.5)	216 (2.5)	16 (2.3)	0.727
Other / unknown	429 (4.6)	410 (4.8)	19 (2.7)	0.009
Ethnicity
Non-Hispanic	8485 (91.4)	7829 (91.2)	656 (94.4)	0.005
Hispanic	742 (8.0)	705 (8.2)	37 (5.3)	0.007
Unknown	56 (0.6)	54 (0.6)	2 (0.3)	0.264
Functionally independent	8800 (94.8)	8167 (95.1)	633 (91.0)	<0.001
Hypertension	3909 (42.1)	3572 (41.6)	337 (48.4)	<0.001
Congestive heart failure	82 (0.9)	76 (0.9)	6 (0.9)	0.951
Ascites	101 (1.1)	92 (1.1)	9 (1.3)	0.586
Disseminated cancer	651 (7.0)	580 (6.8)	71 (10.2)	0.001
Open or infected wound	241 (2.6)	210 (2.4)	31 (4.5)	0.001
Recent weight loss	680 (7.3)	617 (7.2)	63 (9.0)	0.068
Bleeding disorder	338 (3.6)	304 (3.5)	34 (4.9)	0.068
Transfusion >4 units	100 (1.1)	91 (1.1)	9 (1.3)	0.565
SIRS or sepsis
None	8266 (89.0)	7661 (89.2)	605 (86.9)	0.069
SIRS	558 (6.0)	520 (6.1)	38 (5.5)	0.527
Sepsis	299 (3.2)	268 (3.1)	31 (4.5)	0.055
Septic shock	165 (1.8)	143 (86.7)	22 (13.3)	0.004
Dyspnea	986 (10.8)	899 (10.5)	87 (12.5)	0.093
Ventilator dependent >48 h	135 (1.5)	118 (1.4)	17 (2.4)	0.023
Chronic obstructive pulmonary disease	365 (3.9)	322 (3.8)	43 (6.2)	0.002
Acute renal failure	42 (0.5)	36 (0.4)	6 (0.9)	0.094
On dialysis	88 (1.0)	69 (0.8)	19 (2.7)	<0.001

SSI = surgical site infection; ASA = American Society of Anesthesiologists; SD = standard deviation; SIRS = systemic inflammatory response syndrome.

Table 2.

Uni-Variable Analyses of Procedure Characteristics and Surgical Site Infection

	n (%) [Total n = 9,288]	No SSI [n = 8,592]	SSI [n = 696]	p
Procedure characteristics
Any iAE	183 (2.0)	149 (1.7)	34 (4.9)	<0.001
General surgery type
Foregut	1695 (18.3)	1644 (19.1)	51 (7.3)	<0.001
Hepatobiliary	2580 (27.8)	2314 (26.9)	266 (38.2)	<0.001
Intestinal	4212 (45.4)	3852 (44.8)	360 (51.7)	<0.001
Abdominal wall	801 (8.6)	782 (9.1)	19 (2.7)	<0.001
Operation complexity (RVU); mean (SD)	35.3 (23.8)	34.1 (23.0)	49.7 (27.7)	<0.001
1st IQR	3856 (43.4)	3686 (45.2)	170 (24.6)	<0.001
2nd IQR	1696 (19.2)	1580 (19.4)	116 (16.8)	<0.001
3rd IQR	1672 (18.9)	1519 (18.6)	153 (22.1)	<0.001
4th IQR	1624 (18.4)	1371 (16.8)	253 (36.6)	<0.001
Emergency operation	1284 (13.8)	1195 (13.9)	89 (12.8)	0.41
Wound class
Clean	1101 (11.9)	1077 (12.5)	24 (3.5)	<0.001
Clean-contaminated	6308 (67.9)	5814 (67.7)	494 (71.0)	0.072
Contaminated	1156 (12.5)	1064 (12.4)	92 (13.2)	0.521
Dirty-infected	723 (7.8)	637 (7.4)	86 (12.4)	<0.001
Adhesiolysis	1705 (18.4)	1519 (17.7)	186 (26.7)	<0.001
Laparoscopy	4334 (46.7)	4201 (48.9)	133 (19.1)	<0.001
Operation duration, hours; mean (SD)	2.6 (1.9)	2.5 (1.8)	3.6 (2.3)	< 0.001

SSI = surgical site infection; iAE = intra-operative adverse event; RVU = relative value unit; S = standard deviation; IQR = interquartile range.

Uni-variable analyses: Patients with and without SSI

A priori SSI risk factors

Age >60, diabetes mellitus, obesity, smoking, steroid use, and ASA class ≥III were deemed to be a priori risk factors for SSI before the start of our study. Uni-variable analyses of these a priori SSI risk factors and SSI occurrence are displayed in Table 1. Patients in whom SSI developed were more likely to be >60 years old (52.9% vs. 39.5%, p < 0.001), diabetic (18.1% vs. 15.2%, p = 0.04), long-term steroid users (6.2% vs. 4.0%, p = 0.005), and have an ASA class ≥III (47.7% vs. 34.4%, p < 0.0001).

Demographic and pre-operative characteristics

The detailed comparisons of all additional demographic and pre-operative characteristics of patients with and without SSI are reported in Table 1. In summary, patients in whom SSI developed were more likely to have a diagnosis of hypertension (48.4% vs. 41.6%, p < 0.001), disseminated cancer (10.2% vs. 6.8%, p = 0.001), an open/infected wound before operation (4.5% vs. 2.4%, p = 0.001), septic shock (3.1% vs. 1.7%, p = 0.004), ventilator dependence >48 hours (2.4% vs. 1.4%, p = 0.023), chronic obstructive pulmonary disease (6.2% vs. 3.8%, p = 0.002), or be dialysis dependent (2.7% vs. 0.8%, p < 0.001). In addition, patients with SSI were less likely to be functionally independent (91% vs. 95.1%, p < 0.001) compared with those in whom SSI did not develop.

Procedure characteristics

The detailed comparisons of procedure characteristics of patients with and without SSI are reported in Table 2. Patients in whom SSI developed were more likely to have at least one iAE occur during the operation compared with patients in whom SSI did not develop (4.9% vs. 1.7%, p < 0.001). Patients in whom SSI developed were more likely to have undergone a hepatobiliary operation (38.2% vs. 26.9%, p < 0.001), intestinal operation (51.7% vs. 44.8%, p < 0.001), and have an adhesiolysis performed (26.7% vs. 17.7%, p < 0.001). The SSI was also associated with increasingly complex operations (mean RVU 49.7 ± 27.7 vs. 34.1 ± 23.0), dirty wounds (12.4% vs. 7.4%, p < 0.001), and longer operations (mean hours 3.6 ± 2.3 vs. 2.5 ± 1.8, p < 0.001). Patients with a SSI were less likely to have had a laparoscopic approach (19.1% vs. 48.9%, p < 0.001).

Characterization of iAEs

Of the 9,288 patients, 183 had a documented iAE. Table 3 compares the severity, nature, and timing of the iAE in patients with and without SSI. There were 73 major iAEs, which required tissue/organ resection. Patients in whom SSI developed were significantly more likely to have had a major iAE compared with those in whom SSI did not develop (64.7% vs. 34.3%, p < 0.001). The majority of iAEs were injuries to the small bowel (36.1%) or a blood vessel (29.6%). There were no significant differences between the SSI and non-SSI groups in the type of iAE that occurred.

Table 3.

Uni-Variable Analyses of Intra-Operative Adverse Event Characteristics and Surgical Site Infection for Patients Who Had ≥1 Intra-Operative Adverse Event

	n (%) [Total n = 183]	No SSI [n = 149]	SSI [n = 34]	p
iAE characteristics
Severity of iAE
Minor	110 (60.1)	98 (65.8)	12 (35.3)	<0.001
Major	73 (39.9)	51 (34.2)	22 (64.7)	<0.001
Type of iAE
Vascular injury
Minor vessel	12 (6.6)	11 (7.4)	1 (2.9)	0.345
Major vessel	42 (23.0)	35 (23.5)	7 (16.7)	0.717
Bowel injury
Small bowel	66 (36.1)	51 (34.2)	15 (44.1)	0.279
Colon	15 (8.2)	10 (6.7)	5 (14.7)	0.125
Other organ injury
Liver	7 (3.8)	7 (4.7)	0 (0)	0.197
Spleen	15 (8.2)	14 (9.4)	1 (2.9)	0.216
Extrahepatic bile ducts	6 (3.3)	5 (3.7)	1 (2.9)	0.903
Bladder	6 (3.3)	5 (3.7)	1 (2.9)	0.903
Phase of operation that iAE occurred
Access to surgical site	9 (4.9)	8 (5.4)	1 (2.9)	0.555
Dissection	108 (59.0)	89 (59.7)	19 (55.9)	0.680
Retraction	11 (6.0)	11 (7.4)	0 (0)	0.102
Resection/reconstruction	36 (19.7)	30 (20.1)	6 (17.7)	0.742
Unclear	19 (10.4)	11 (7.4)	8 (23.5)	0.005

SSI = surgical site infection; iAE = intra-operative adverse event.

Multivariable analyses: Independent impact of intra-operative adverse events on the risk of SSI

Each multivariable analysis was performed independently by adjusting for significant pre-operative characteristics, in addition to case complexity, wound classification, and minimally invasive approach.

Severity of intra-operative adverse events as a risk factor for SSI by type

Table 4 summarizes the multi-variable analyses performed to identify the impact that severity of an iAE has on developing each type of SSI. In summary, the occurrence of any iAE was associated independently with an increased risk of any SSI (OR = 1.67; 95% CI, 1.11–2.52; p = 0.013). Not surprisingly, major iAEs were associated with the highest risk for SSI (OR = 3.39; 95% CI, 1.97–5.85; p < 0.001). In terms of specific types of SSI, the occurrence of any iAE increased the risk of organ space infections (OR = 1.81 95% CI, 1.07–3.05; p = 0.027], and the risk was again greater for major iAEs (OR = 3.99; 95% CI, 2.12–7.52; p < 0.001]. The occurrence of an iAE, regardless of severity, did not associate independently with the development of superficial or deep incisional SSIs.

Table 4.

Multi–Variable Analysis: Severity of Intra-Operative Adverse Event as a Risk Factor for Surgical Site Infection by Type

	Independent variable
	Any iAE (n = 183)		Major iAE (n = 73)
Dependent outcomes	OR (95% CI)	p	OR (95% CI)	p
Any surgical site infection	1.67 (1.11–2.52)	0.013^†	3.39 (1.97–5.85)	<0.001^†
Superficial infection	1.29 (0.7–2.37)	0.411	1.80 (0.76–4.25)	0.181
Deep incisional infection	2.72 (0.61–12.25)	0.191	3.33 (0.42–26.4)	0.255
Organ space infection	1.81 (1.07–3.05)	0.027^†	3.99 (2.12–7.52)	<0.001^†

iAE = intra-operative adverse event; OR = odds ratio; CI = confidence interval.

†

= Significant

Impact of any intra-operative adverse event on the risk of SSI

Figure 2 summarizes the multi-variable analyses performed to identify the impact that any iAE has on development of any SSI. For each pre-determined a priori SSI risk factor, the patients were divided into high-risk and low-risk patient cohorts, and independent analyses were performed. Interestingly, the occurrence of an iAE was correlated with increased SSI rate in the low-risk but not the high-risk patient populations. Specifically, iAEs increased SSI in patients younger than age 60 (OR = 2.69; 95% CI, 1.55–4.67; p < 0.001], those without diabetes mellitus (OR = 1.64; 95% CI, 1.04–2.58; p = 0.034), non-obese (OR = 2.9; 95% CI, 1.81–4.66; p < 0.001), non-smokers (OR = 1.67; 95% CI, 1.08–2.60; p = 0.022), with no steroid use (OR = 1.73; 95% CI, 1.15–2.6; p = 0.008), and with ASA class <III (OR = 2.26; 95% CI, 1.31–3.87; p = 0.003).

FIG. 2.

Multi-variable analysis: Any intra-operative adverse event as a risk factor for surgical site infection for low risk patient cohorts and high risk patient cohorts. ASA = American Society of Anesthesiologists.

Impact of major intra-operative adverse events on the risk of SSI

Figure 3 summarizes the multi-variable analyses that were performed to identify the impact that major iAEs have on development of SSI. Analyses of patient cohorts by a priori risk factors revealed an independent association of major iAEs with SSI in both high-risk and low-risk patients when considering age, obesity, smoking, and ASA class. This association, however, was only significant for low-risk patients when considering diabetes mellitus (non-diabetics OR = 3.71; 95% CI, 2.05–6.71; p < 0. 001) and steroid use (no steroids OR = 3.74; 95% CI, 2.18–6.43; p < 0.001).

FIG. 3.

Multi-variable analysis: Major intra-operative adverse events as a risk factor for surgical site infection for low risk patient cohorts and high risk patient cohort. ASA = American Society of Anesthesiologists.

Discussion

In this study, we provide a detailed analysis of the independent effect of iAEs on SSI in abdominal operations. The occurrence of an iAE increases the risk of post-operative SSI, specifically organ/space infections. Further, as the severity of the iAE increases, the risk of SSI increases. Most interestingly, the risk of SSI developing after the occurrence of any iAE was most significant for the patients thought to be low risk based on their pre-operative co-morbidities. In that patient population, an iAE, especially a major iAE, shifts them from a low risk to a high risk for SSI.

With an annual cost of $3.5–$10 billion, SSIs provide more financial burden to our healthcare systems than all other hospital-acquired infections [1,31]. There has been a considerable amount of research into the prevention, detection, and management of SSI. The most current literature was included recently in the updated ACS and Surgical Infection Society guidelines for SSI [17]. These guidelines discuss procedure-related risk factors for SSI to aid in early detection. While the guidelines do include factors that could be considered downstream effects of an iAE (i.e., blood transfusion or prolonged operative time), iAEs are not considered independently given the lack of data regarding their association with SSI.

The study of iAEs has expanded in recent years because of the increasing emphasis on post-surgical outcomes. Ernest Amory Codman [32] was perhaps the first to record iAEs in the early 20th century and to follow systematically to an “end-result” all of his patients including those with intra-operative complications. Since this early description, various groups have attempted to study specific aspects of iAEs [33 –43]. In regard to surgical infections, the majority of the literature before our study is focused on particular adverse events or operations, which limits generalizability [44,45]. Kin and colleagues [45] studied the incidence of accidental puncture or laceration during 2,897 colon resections and the association with post-operative outcomes. They found no significant association between iAEs and wound infections (superficial and deep incisional). While only 1.6% of patients who had a serosal tear had an intra-abdominal abscess develop, however, this complication increased to 6.9% and 10% for those who had inadvertent enterotomies or extra-intestinal injuries, respectively (p = 0.03).

Additional studies have demonstrated that incisional and organ/space infections may have unique risk factors [46]. While the study by Kin and colleagues [45] is limited to colon resections, it corroborates our finding that iAEs are associated with an increased risk of organ/space infections. In addition, although they did not utilize a validated system to classify iAE severity, it can be implied that they detected a direct relationship between iAE severity and the risk of abscess formation. Taken together, these data suggest that iAEs may be an independent risk factor for organ/space infections and that the risk increases with iAE severity.

Our previous work has demonstrated that 92% of iAEs are recognized and addressed at the time of the surgical procedure without the need for re-operation [19]. Despite this early detection, these patients still have an SSI develop at a significantly increased rate. We must consider the interventions that should be taken intra-operatively to decrease the rate of infection. First and foremost, the prevention of SSI begins before the operation by strict adherence to the SCIP measures [18,19]. They include appropriate timing, dosing, and choice of peri-operative antibiotic agents, hair removal method, intra-operative normothermia, and glucose control. Additional studies have found that optimizing tissue oxygenation and utilization of surgical safety checklists also reduce the risk of SSI [17,47 –50].

Perhaps an important intervention that can be made by the surgeon once an iAE is encountered, aside from appropriate repair/reconstruction, is enacting an “intra-operative time out.” This would allow dedicated time during the operation for surgery, anesthesia, and the nursing staff to discuss the injury and ensure each of the aforementioned quality indicators that reduce SSI continue to be followed. Specifically, it should be discussed whether it is necessary to re-dose prophylactic peri-operative antibiotic agents based on the duration of the operation or estimated blood loss, both of which may be increased as a result of the injury [51,52]. At present, it is not recommended to continue prophylactic antibiotic agents beyond the time of skin closure. This intervention, similar to the routine use of closed suction drainage systems, has not been shown to decrease the rate of SSI [17,53,54].

In our study, we attempted to determine further how the occurrence of an iAE impacted patients with known pre-operative risk factors for SSI. Surprisingly, the occurrence of any iAE was associated with an increase rate of SSI only in the lower-risk patients (e.g., non-diabetics, non-steroid users), compared with their higher-risk counterparts (e.g., diabetics, steroid users). This finding, while initially counterintuitive, is likely attributed to the baseline low rate of SSI in patients lacking a priori risk factors. For higher-risk patients, however, the effect is not additive and does not relatively increase their already elevated risk of SSI at baseline. Therefore, the occurrence of any iAE will essentially convert a low-risk patient into a higher-risk patient.

Our study has a few limitations. First, we were not able to determine whether appropriate antibiotic prophylaxis was given during each operation, which may confound our data. Second, the false negative rate of the APL algorithm used to screen our linked database for intra-operative occurrences ranges from 2% to 6% [25,55,56]. As a result, a small subset of patients may have been misclassified as not having an iAE. The positive predictive value of our PSI #15 is 85% to 91%. Operative reports were reviewed manually, however, to exclude false positives [25,56,57]. Third, our review of operative notes is dependent on the quality of the dictated operative report, which is inherently variable. As a consequence, minor iAEs that may not have been dictated by particular providers may be underreported and therefore misclassified in our study. This limitation highlights the pressing need for improving methods of detecting all intra-operative variables (including iAEs). Finally, as is the case with studies of this nature, we can only prove association but not causation between iAEs and SSI.

Conclusions

The iAEs are associated independently with a significant increase in SSI—specifically organ space infections—particularly in patients with less pre-existing risk factors. Major iAEs that require tissue resection or reconstruction are associated with a higher risk of SSI compared with minor iAEs. Future research and quality improvement efforts should focus on prevention of iAEs, mitigation of their adverse effects, and incorporation of iAEs into prediction models for post-operative SSI.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Anderson

, Podgorny

, Berrios-Torres

, et al. Strategies to prevent surgical site infections in acute care hospitals: 2014 update. Infect Control Hosp Epidemiol, 2014; 35(Suppl 2):S66–S88.

Barie

. Surgical site infections: Epidemiology and prevention. Surg Infect, 2002; 3(Suppl 1):S9–S21.

Ejaz

, Schmidt

, Johnston

, et al. Risk factors and prediction model for inpatient surgical site infection after major abdominal surgery. J Surg Res, 2017; 217:153–159.

Pinkney

, Calvert

, Bartlett

, et al. Impact of wound edge protection devices on surgical site infection after laparotomy: Multicentre randomised controlled trial (ROSSINI Trial). BMJ, 2013; 347:f4305.

Astagneau

, Rioux

, Golliot

, et al. Morbidity and mortality associated with surgical site infections: Results from the 1997–1999 INCISO surveillance. J Hosp Infect, 2001; 48:267–274.

GlobalSurg Collaborative. Determining the worldwide epidemiology of surgical site infections after gastrointestinal resection surgery: Protocol for a multicentre, international, prospective cohort study (GlobalSurg 2). BMJ Open, 2016; 7:e012150.

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.

Merle

, Germain

, Chamouni

, et al. Assessment of prolonged hospital stay attributable to surgical site infections using appropriateness evaluation protocol. Am J Infect Control, 2000; 28:109–115.

Shepard

, Ward

, Milstone

, et al. Financial impact of surgical site infections on hospitals: The hospital management perspective. JAMA Surg, 2013; 148:907–914.

10.

Herwaldt

, Cullen

, Scholz

, et al. A prospective study of outcomes, healthcare resource utilization, and costs associated with postoperative nosocomial infections. Infect Control Hosp Epidemiol, 2006; 27:1291–1298.

11.

Badia

, Casey

, Petrosillo

, et al. Impact of surgical site infection on healthcare costs and patient outcomes: A systematic review in six European countries. J Hosp Infect, 2017; 96:1–15.

12.

World Health Organization. Global Guidelines for the Prevention of Surgical Site Infection. Geneva: World Health Organization, 2016.

13.

Hall

, DeFrances

C.J.

, Williams

S.N.

, Golosinskiy

, Swartzman

. National Health Statistics Report. National Hospital Discharge Survey: 2007 Summary. 2010. www.cdc.gov/nchs/data/nhsr/nhsr029.pdf. (Last accessed 3/15/2018).

14.

Winfield

, Reese

, Bochicchio

, et al. Obesity and the risk for surgical site infection in abdominal surgery. Am Surg, 2016; 82:331–336.

15.

Buggy

. Can anaesthetic management influence surgical-wound healing?. Lancet, 2000; 356:355–357.

16.

Cheadle

. Risk factors for surgical site infection. Surg Infect, 2006; 7(Suppl 1):S7–S11.

17.

Ban

, Minei

, Laronga

, et al. American College of Surgeons and Surgical Infection Society: Surgical Site Infection Guidelines, 2016 Update. J Am Coll Surg, 2017; 224:59–74.

18.

Rosenberger

, Politano

, Sawyer

. The surgical care improvement project and prevention of post-operative infection, including surgical site infection. Surg Infect, 2011; 12:163–168.

19.

Bohnen

, Mavros

, Ramly

, et al. Intraoperative adverse events in abdominal surgery: What happens in the operating room does not stay in the operating room. Ann Surg, 2017; 265:1119–1125.

20.

Kaafarani

, Mavros

, Hwabejire

, et al. Derivation and validation of a novel severity classification for intraoperative adverse events. J Am Coll Surg, 2014; 218:1120–1128.

21.

Mavros

, Bohnen

, Ramly

, et al. Intraoperative adverse events: Risk adjustment for procedure complexity and presence of adhesions is crucial. J Am Coll Surg, 2015; 221:345–353.

22.

Mavros

, Velmahos

, Larentzakis

, et al. Opening Pandora's box: Understanding the nature, patterns, and 30-day outcomes of intraoperative adverse events. Am J Surg, 2014; 208:626–631.

23.

American College of Surgeons National Surgical Quality Improvement Program. User Guide for the 2015 ACS-NSQIP Participant Use Data File. 2016. www.facs.org/∼/media/files/quality%20programs/nsqip/nsqip_puf_user_guide_2015.ashx. (Last accessed 4/1/2017).

24.

Horan

, Gaynes

, Martone

, et al. CDC definitions of nosocomial surgical site infections, 1992: A modification of CDC definitions of surgical wound infections. Infect Control Hosp Epidemiol, 1992; 13:606–608.

25.

Kaafarani

, Borzecki

, Itani

, et al. Validity of selected Patient Safety Indicators: Opportunities and concerns. J Am Coll Surg, 2011; 212:924–934.

26.

Davenport

, Henderson

, Khuri

, Mentzer

Jr . Preoperative risk factors and surgical complexity are more predictive of costs than postoperative complications: A case study using the National Surgical Quality Improvement Program (NSQIP) database. Ann Surg, 2005; 242:463–468.

27.

Henderson

, Daley

. Design and statistical methodology of the National Surgical Quality Improvement Program: Why is it what it is?. Am J Surg, 2009; 198:S19–S27.

28.

Johnson

, Arozullah

, Neumayer

, et al. Multivariable predictors of postoperative respiratory failure after general and vascular surgery: Results from the patient safety in surgery study. J Am Coll Surg, 2007; 204:1188–1198.

29.

Neumayer

, Hosokawa

, Itani

, et al. Multivariable predictors of postoperative surgical site infection after general and vascular surgery: Results from the patient safety in surgery study. J Am Coll Surg, 2007; 204:1178–1187.

30.

Virani

, Michaelson

, Hutter

, et al. Morbidity and mortality after liver resection: Results of the patient safety in surgery study. J Am Coll Surg, 2007; 204:1284–1292.

31.

Magill

, Edwards

, Bamberg

, et al. Multistate point-prevalence survey of health care-associated infections. N Engl J Med, 2014; 370:1198–1208.

32.

Codman

. The classic: A study in hospital efficiency: As demonstrated by the case report of first five years of private hospital. Clin Orthop Relat Res, 2013; 471:1778–1783.

33.

Brennan

, Leape

, Laird

, et al. Incidence of adverse events and negligence in hospitalized patients: Results of the Harvard Medical Practice Study I. 1991. Qual Saf Health Care, 2004; 13:145–151.

34.

Couch

, Tilney

, Rayner

, Moore

. The high cost of low-frequency events: The anatomy and economics of surgical mishaps. N Engl J Med, 1981; 304:634–637.

35.

Gawande

, Thomas

, Zinner

, Brennan

. The incidence and nature of surgical adverse events in Colorado and Utah in 1992. Surgery, 1999; 126:66–75.

36.

Wojcik

, Lee

, Peponis

, et al. Do not blame the resident: The impact of surgeon and surgical trainee experience on the occurrence of intraoperative adverse events (iAEs) in abdominal surgery. J Surg Educ, 2018; 75:e156–e167.

37.

Peponis

, Baekgaard

, Bohnen

, et al. Are surgeons reluctant to accurately report intraoperative adverse events? A prospective study of 1,989 patients. Surgery, 2018; 164:525–529.

38.

Eskesen

, Peponis

, Saillant

, et al. Operating at night does not increase the risk of intraoperative adverse events. Am J Surg, 2018; 216:19–24.

39.

Han

, Bohnen

, Peponis

, et al. The surgeon as the second victim? Results of the Boston Intraoperative Adverse Events Surgeons' Attitude (BISA) Study. J Am Coll Surg, 2017; 224:1048–1056.

40.

Nandan

, Bohnen

, Chang

, et al. The impact of major intraoperative adverse events on hospital readmissions. Am J Surg, 2017; 213:10–17.

41.

Ramly

, Bohnen

, Farhat

, et al. The nature, patterns, clinical outcomes, and financial impact of intraoperative adverse events in emergency surgery. Am J Surg, 2016; 212:16–23.

42.

Ramly

, Larentzakis

, Bohnen

, et al. The financial impact of intraoperative adverse events in abdominal surgery. Surgery, 2015; 158:1382–1388.

43.

Kaafarani

, Velmahos

. Intraoperative adverse events: The neglected quality indicator of surgical care?. Surgery, 2015; 157:6–7.

44.

Garteiz

, Guzman

, Alonso

, et al. Gallbladder rupture during laparoscopic cholecystectomy: Does it have an effect on postoperative morbidity?. Surg Laparosc Endosc Percutan Tech, 1999; 9:263–266.

45.

Kin

, Snyder

, Kiran

, et al. Accidental puncture or laceration in colorectal surgery: A quality indicator or a complexity measure?. Dis Colon Rectum, 2013; 56:219–225.

46.

, Stein

, Trencheva

, et al. Differing risk factors for incisional and organ/space surgical site infections following abdominal colorectal surgery. Dis Colon Rectum, 2011; 54:818–825.

47.

Stulberg

, Delaney

, Neuhauser

, et al. Adherence to surgical care improvement project measures and the association with postoperative infections. JAMA, 2010; 303:2479–2485.

48.

Qadan

, Akca

, Mahid

, et al. Perioperative supplemental oxygen therapy and surgical site infection: A meta-analysis of randomized controlled trials. Arch Surg, 2009; 144:359–366.

49.

Qadan

, Battista

, Gardner

, et al. Oxygen and surgical site infection: A study of underlying immunologic mechanisms. Anesthesiology, 2010; 113:369–377.

50.

Meyhoff

, Wetterslev

, Jorgensen

, Rasmussen

. High perioperative inspiratory oxygen fraction to prevent anastomotic dehiscence. J Gastrointest Surg, 2013; 17:842.

51.

Bergs

, Hellings

, Cleemput

, et al. Systematic review and meta-analysis of the effect of the World Health Organization surgical safety checklist on postoperative complications. Br J Surg, 2014; 101:150–158.

52.

American College of Surgeons. American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) Surgical Risk Calculator. www.riskcalculator.facs.org/RiskCalculator/. (Last accessed February 15, 2017).

53.

, Barie

, Stein

, et al. Antibiotic regimen and the timing of prophylaxis are important for reducing surgical site infection after elective abdominal colorectal surgery. Surg Infect, 2011; 12:255–260.

54.

Reiffel

, Barie

, Spector

. A multi-disciplinary review of the potential association between closed-suction drains and surgical site infection. Surg Infect, 2013; 14:244–269.

55.

Guzman-Valdivia

. Routine administration of antibiotics to patients suffering accidental gallbladder perforation during laparoscopic cholecystectomy is not necessary. Surg Laparosc Endosc Percutan Tech, 2008; 18:547–550.

56.

Utter

, Zrelak

, Baron

, et al. Positive predictive value of the AHRQ accidental puncture or laceration patient safety indicator. Ann Surg, 2009; 250:1041–1045.

57.

Rosen

, Itani

, Cevasco

, et al. Validating the patient safety indicators in the Veterans Health Administration: Do they accurately identify true safety events?. Med Care, 2012; 50:74–85.