Improving prediction of surgical site infection risk with multilevel modeling

The Accuracy of the Learning-Curve Cumulative Sum Method in Assessing Brachial Plexus Block Competency

BACKGROUND

The learning-curve cumulative sum method (LC-CUSUM) and its risk-adjusted form (RA-LC-CUSUM) have been proposed as performance-monitoring methods to assess competency during the learning phase of procedural skills. However, scarce data exist about the method's accuracy. This study aimed to compare the accuracy of LC-CUSUM forms using historical data consisting of sequences of successes and failures in brachial plexus blocks (BPBs) performed by anesthesia residents.

METHODS

Using historical data from 1713 BPB performed by 32 anesthesia residents, individual learning curves were constructed using the LC-CUSUM and RA-LC-CUSUM methods. A multilevel logistic regression model predicted the procedure-specific risk of failure incorporated in the RA-LC-CUSUM calculations. Competency was defined as a maximum 15% cumulative failure rate and was used as the reference for determining the accuracy of both methods.

RESULTS

According to the LC-CUSUM method, 22 residents (84.61%) attained competency after a median of 18.5 blocks (interquartile range [IQR], 14-23), while the RA-LC-CUSUM assigned competency to 20 residents (76.92%) after a median of 17.5 blocks (IQR, 14-25, P = .001). The median failure rate at reaching competency was 6.5% (4%-9.75%) under the LC-CUSUM and 6.5% (4%-9%) for the RA-LC-CUSUM method (P = .37). The sensitivity of the LC-CUSUM (85%; 95% confidence interval [CI], 71%-98%) was similar to the RA-LC-CUSUM method (77%; 95% CI, 61%-93%; P = .15). Identical specificity values were found for both methods (67%; 95% CI, 29%-100%, P = 1).

CONCLUSIONS

The LC-CUSUM and RA-LC-CUSUM methods were associated with substantial false-positive and false-negative rates. Also, small lower limits for the 95% CIs around the accuracy measures were observed, indicating that the methods may be inaccurate for high-stakes decisions about resident competency at BPBs.

Algerian postcaesarean surgical site infections: A cross-sectional investigation of the epidemiology, bacteriology, and antibiotic resistance profile

BACKGROUND

Surgical site infections (SSIs) are one of the most common health care-associated infections in low and middle-income countries. The aims of this cross-sectional descriptive study were to estimate the frequency of postcaesarean infection with associated clinical characteristics and the antibiotic resistance profile of bacterial isolates.

METHODS

Patients who underwent a cesarean section at the obstetrics and gynecology department of the hospital in Annaba, Algeria were included. Each woman was followed postoperatively for 30 days and sociodemographic data were collected. Culture-based microbiological methods were used to identify the causative bacteria and determine their antibiotic resistance phenotype and molecular characterization.

RESULTS

Among 1,810 patients, we recorded 36 (1.9%) SSIs. Most patients had undergone an emergency delivery (75%) and low educational level (72.2%). The most frequent maternal pathologies were Body Mass Index ≥ 30 (63.9%), scarred uteri (58.3%), anemia (55.6%), and an American Society of Anaesthesiologists score between II and III (33.3%). Of the 43 bacteria isolated, Enterobacteriaceae were the most frequent (62.8%), predominated by Escherichia coli strains (43.5%), a majority of which were extended-spectrum β-lactamases carriers (62.9%). Although gram-positive cocci were less frequent (37.2%), a majority of Enterococcus faecalis (56.2%) were observed and 2 strains of vancomycin-resistant Enterococcus faecium harboring the vanA gene were identified.

CONCLUSIONS

Extensive surveillance of at-risk populations should be integrated to prevent the occurrence of SSIs.

From the local disparities to national realities: Mapping and multilevel modeling of catastrophic health expenditure in Bangladesh using HIES 2016

In developing nations, catastrophic health expenditures have become an all-too-common occurrence, threatening to push households into impoverishment and poverty. By analyzing the Household Income and Expenditure Survey 2016, which features a sample of 46,080 households, this study provides a comprehensive district-by-district analysis of the variation in household catastrophic health expenditures and related factors. The study utilizes a multilevel logistic regression model, which considers both fixed and random effects to identify factors associated with catastrophic health expenditure. The findings of the study indicate that districts located in the eastern and southern regions are at a significantly higher risk of experiencing catastrophic health expenditures. A potential explanation for this trend may be attributed to the high prevalence of chronic diseases in these districts, as well as their economic conditions. The presence of chronic diseases (AOR 5.45 with 95% CI: 5.14, 5.77), presence of old age person (AOR 1.50 with 95% CI: 1.39, 1.61), place of residence (AOR 1.40 with 95% CI: 1.14, 1.73) are found to be highly associated factors. Additionally, the study reveals that the thresholds used to define catastrophic health expenditures exhibit substantial variation across different regions, and differ remarkably from the threshold established by the WHO. On average, the thresholds are 23.12% of nonfood expenditure and 12.14% of total expenditure. In light of these findings, this study offers important insights for policymakers and stakeholders working towards achieving universal health coverage and sustainable development goals in Bangladesh.

Effect of preoperative hospital stay on surgical site infection in Chinese cranial neurosurgery

OBJECTIVE

Surgical site infection(SSI)after neurosurgical procedure can be devastating. Delayed hospital stay has been identified as a potentially modifiable driver of SSI in general surgery patients. However, the relationship between preoperative length of stay and SSI has not been quantified previously in neurosurgery. This study aimed to clarify the association.

DESIGN

A Cohort study based on STROBE checklist.

METHOD

This observational study focused on cranial neurosurgery patients at a tertiary referral centers in China. Data collection from hospital information system conducted between 1 January 2016 and 31 December 2016 was used to examine the results of interest (n = 600). Logistic regression analysis explored association between preoperative length of stay and SSI, adjusting for potential confounders.

RESULTS

Overall SSI prevalence was 10.8% and was significantly higher in the longer preoperative length of stay group. Besides preoperative length of stay, American Society of Anesthesiologists score, type of surgery, gross blood loss also significantly associated with SSI prevalence. Compared with 1 to 2 days, longer preoperative length of stay was associated with increased SSI prevalence after adjustment for confounders (3 to 4 days: odds ratio[OR], 0.975[95%CI, 0.417 to 2.281]; 5 to 6 days: OR, 2.830[95%CI, 1.092 to 7.332]; 7 or more days: OR, 4.039[95%CI, 1.164 to 14.015]; P for trend < 0.001). On the other hand, we found a positive association between preoperative length of stay to deep/space-organ SSI (OR = 1.404; 95% CI: 1.148 to 1.717; P for trend < 0.001), which was higher than superficial SSI (OR = 1.242; 95% CI: 0.835 to1.848; P for trend= 0.062).

CONCLUSIONS

In a cohort of patients from a single center retrospective surgical registry, a longer preoperative length of stay was associated with a higher incidence of cranial neurosurgical SSI. There is room for improvement in preoperative length of stay. This can be used for hospital management and to stratify patients with regard to SSI risk.

Hospital context in surgical site infection following colorectal surgery: a multi-level logistic regression analysis

BACKGROUND

Surgical site infections (SSIs) are associated with poor health outcomes. Their incidence is highest after colorectal surgery, with little improvement in recent years. The role of hospital characteristics is undetermined.

AIM

To investigate whether SSI incidence after colorectal surgery varies between hospitals, and whether such variance may be explained by hospital characteristics.

METHODS

Data were retrieved from the electronic platform of the Directorate General of Health, from 2015 to 2019. Hospital characteristics were retrieved from publicly available data on the Portuguese public administration. Analysis considered a two-level hierarchical data structure, with individuals clustered in hospitals. To avoid overfitting, no models were built with more than one hospital characteristic. Cluster-level associations are presented through median odds ratio (MOR) and intraclass cluster coefficient (ICC). Beta coefficients were used to assess the contextual effects.

FINDINGS

A total of 11,219 procedures from 18 hospitals were included. The incidence of SSI was 16.8%. The ICC for the null model was 0.09. Procedural variables explained 25% of the variance, and hospital dimension explained another 17%. More than 50% of SSI variance remains unaccounted for. After adjustment, heterogeneity between hospitals (MOR: 1.51; ICC: 0.05) was still found. No hospital characteristic was significantly associated with SSI.

CONCLUSION

Procedural variables and hospital dimension explain almost half of SSI variance and should be taken into account when implementing prevention strategies. Future research should focus on compliance with preventive bundles and other process indicators in hospitals with significantly less SSI in colorectal surgery.

Clinically relevant features for predicting the severity of surgical site infections

Surgical site infections are hospital-acquired infections resulting in severe risk for patients and significantly increased costs for healthcare providers. In this work, we show how to leverage irregularly sampled preoperative blood tests to predict, on the day of surgery, a future surgical site infection and its severity. Our dataset is extracted from the electronic health records of patients who underwent gastrointestinal surgery and developed either deep, shallow or no infection. We represent the patients using the concentrations of fourteen common blood components collected over the four weeks preceding the surgery partitioned into six time windows. A gradient boosting based classifier trained on our new set of features reports, respectively, an AUROC of 0:991 and 0:937 at predicting a postoperative infection and the severity thereof. Further analyses support the clinical relevance of our approach as the most important features describe the nutritional status and the liver function over the two weeks prior to surgery.

Prevalence of surgical site infection after orthopaedic surgery with two types of drainage at three public hospitals in Iran

INTRODUCTION

Orthopaedic surgical site infections (SSIs) are among the most important and prevalent because implanted devices are used in such surgeries which increase the risk of infection. This study aimed to examine the incidence of infection in orthopaedic surgeries and related factors in a group of public hospitals in Iran.

METHODS

This analytical cross-sectional study was condcuted in 2018. Data were collected about the incidence of infection and related factors including use/non-use of drains and well as types of drain . The presence of SSIs were determined using a researcher-devised Wound Infection Checklist (WIC) and microbial cultures.

RESULTS

Of 110 included patients undergoing orthopaedic surgeries, 18.2% had an infection. The highest rate of infection was associated with lower extremity surgeries following tibia and fibula fractures. The incidence of infection among individuals under the age of 35 years was also higher. Emergency procedures demonstrated a higher incidence of infection. No significant relationship was observed between use of wound drains and development of SSIs.

CONCLUSION

Reducing the incidence of orthopaedic infection can be achieved by focusing more attention on open fractures occurring in the lower extremity. Use of surgical drains for the sole purpose of reducing the rate of infection is not effective. Accordingly, infection provention protocols should be implemented in orthopaedic surgery units to control and reduce rates of infection.

Risk Factors Associated with Surgical Site Infections: A Retrospective Report from a Developing Country

Introduction

Any infection occurring at the site of a surgical incision superficially or deep within the fascia, within 30 days of a surgical procedure is termed as a surgical site infection (SSI). Due to limited resources, non-adherence to infection control guidelines and substandard sterilization practices, the incidence is higher in developing countries. The aim of this study is to estimate the incidence of surgical site infections in general surgeries at a tertiary care hospital in Pakistan and identify the predisposing risk factors.

Methods

This was a retrospective analysis that included all surgical records from June 1, 2018, to December 31, 2018. After exclusion, 882 records were included. The incidence of SSI and predisposing risk factors were noted. Data were entered and analyzed using SPSS v. 22.0 (IBM Corp, Armonk, NY, US).

Results

The incidence of SSI was 8.84% (n=78). SSIs were more common in older participants (11.4% vs. 6.4%; p=0.009), in patients with more than 24 hour of preoperative hospital stay (11.2% vs. 64%; p=0.013), in procedures of longer duration (1.53 ± 0.35 vs 2.57 ± 0.17; p<0.0001), and in emergency surgeries (19.2% vs. 7.5%; p=0.0001). The combined incidence of SSIs in American Society of Anesthesiologists (ASA) index III and above was 37 (47.4%) and that in I and II was 41 (52.6%) (p<0.00001).

Conclusion

This study has revealed a very high incidence of surgical site infections. These infections are more common in elderly patients, patients undergoing emergency surgeries, those with longer preoperative hospital stay and longer surgical duration, and patients with a high ASA index.

Using the Electronic Health Record Data in Real Time and Predictive Analytics to Prevent Hospital-Acquired Postoperative/Surgical Site Infections

Of critical concern to hospitals today is the prevention of postoperative (surgical site) infections that often result in increased lengths of stays for patients, increased resource demands and costs, loss of public trust and lawsuits, and needless pain and suffering for patients and their families. While all surgical patients have the potential to develop a postoperative infection, the main challenge is to identify key risk factors (both patient centered and operational) through an electronic early-warning system to reduce the likelihood of a postoperative infection from occurring. Currently, most postoperative infection risk prevention practices encompass limited use of informatics technologies or do not maximize the potential benefits. In addition, from a research perspective, there has been more focus on extrapolating electronically housed data (eg, from progress notes, operative notes, laboratory, pharmacy, radiology) retrospectively to describe poor patient outcomes for benchmarking purposes (revealing poor results and opportunities for improvement) rather than using similar sources of real-time data to prevent poor patient outcomes from occurring. This article proposes that standardized indicators, both patient centered and operational, linked to the patient's electronic health record could allow for implementation of 24/7, "real-time" monitoring/surveillance to implement well-timed preventive interventions scaled to each patient and facility to assist caregivers in reducing the numbers of postoperative infections and improve the overall quality and costs of patient care.

Patient autonomy and disclosure of material information about hospital-acquired infections

Hospital-acquired infections are nowadays a major health care problem worldwide. The morbidity and mortality associated with them are highest in intensive care units, but their effects are identifiable in virtually any medical department. Information about hospital-acquired infections, especially about their preventive measures, are rarely presented nowadays in a correct fashion to patients. This article aims to present, in a structured manner, the theoretical and practical aspects related to disclosure of hospital-acquired infections-related information to patients and its importance in preventing their spread. We will analyze both the conceptual framework for disclosing medical information related to nosocomial infections (autonomy, veracity, social justice, the principle of double effect, the precautionary principle, and nonmaleficence) and the practicalities regarding the disclosure of proper information to patients.

Application of antimicrobial drugs in perioperative surgical incision

Infection in surgical incision often results in poor wound healing, and one of the main factors for wound infection is the use of antimicrobial agents. Rational use of antibiotics is one of the key factors to prevent incision infection in general surgery. The number of current clinical studies on antibiotic use before and during surgery is greater than that of systematic studies on antibiotic use after surgery. For the rational use of antibiotics and improvement of wound healing rate, researchers around the world have gradually focused on the use of antibiotics after surgery. Despite the familiarity on the concept of “rational use of antibiotics”, few clear and systematic studies were conducted to elucidate the effect of different antibiotics on wound healing. Therefore, this review focuses on the use of different types of antimicrobial agents in surgical wounds.

Incidence and risk factors for surgical site infection in general surgeries

Objective:

to estimate the incidence of surgical site infection in general surgeries at a large Brazilian hospital while identifying risk factors and prevalent microorganisms.

Method:

non-concurrent cohort study with 16,882 information of patients undergoing general surgery from 2008 to 2011. Data were analyzed by descriptive, bivariate and multivariate analysis.

Results:

the incidence of surgical site infection was 3.4%. The risk factors associated with surgical site infection were: length of preoperative hospital stay more than 24 hours; duration of surgery in hours; wound class clean-contaminated, contaminated and dirty/infected; and ASA index classified into ASA II, III and IV/V. Staphyloccocus aureus and Escherichia coli were identified.

Conclusion:

the incidence was lower than that found in the national studies on general surgeries. These risk factors corroborate those presented by the National Nosocomial Infection Surveillance System Risk Index, by the addition of the length of preoperative hospital stay. The identification of the actual incidence of surgical site infection in general surgeries and associated risk factors may support the actions of the health team in order to minimize the complications caused by surgical site infection.

Development of a surgical site infection prediction model in orthopaedic trauma: The Denver Health Model

BACKGROUND

The CDC's National Healthcare Safety Network's (NHSN) current risk adjustment model for surgical site infections (SSI) following open reduction internal fixation (ORIF) of long bone fractures is a suboptimal predictor of risk. We hypothesized that by including variables known to be associated with SSI following ORIF, we would develop a model that would increase the accuracy and predictability of SSI risk.

METHODS

Patients who underwent ORIF of a long bone between January 1, 2012 and December 31, 2014 were included in the study (n=1543). Patient risk factors, injury risk factors and perioperative risk factors were considered in the development of this model. We developed a risk prediction model for SSI following ORIF and then applied this to a new dataset of ORIF to determine the expected number of infections. This was compared to the expected number of infections calculated using the NHSN risk adjusted model.

RESULTS

The final multivariate model included age (odds ratio: 1.02, p-value<0.001, 95% confidence interval: 1.00-1.04), lower leg fracture (2.63, 0.004, 1.40-4.93), open fracture (1.87, 0.07, 0.93-3.76), American Society of Anesthesiologists (ASA) (2.09, 0.02, 1.07-4.08) and history of methicillin-resistant Staphylococcus aureus (MRSA), which was the most important predictor of infection (7.20, <0.001, 2.61-19.85). The c-index was 0.74 compared to 0.65 for the NHSN model, indicating that our model more accurate in estimating infection risk. When the developed model was used to predict the number of expected infections on a new dataset from 2015, 36.3 SSI were expected compared to 5.7 calculated by the NHSN model.

CONCLUSIONS

The model that was developed uses five easily identifiable risk factors that result in a more accurate prediction of infection at our facility than the currently used model.

LEVEL OF EVIDENCE

Prognostic and epidemiologic study, level III.

A Prognostic Model of Surgical Site Infection Using Daily Clinical Wound Assessment

BACKGROUND

Surgical site infection (SSI) remains a common, costly, and morbid health care-associated infection. Early detection can improve outcomes, yet previous risk models consider only baseline risk factors (BF) not incorporating a proximate and timely data source-the wound itself. We hypothesize that incorporation of daily wound assessment improves the accuracy of SSI identification compared with traditional BF alone.

STUDY DESIGN

A prospective cohort of 1,000 post open abdominal surgery patients at an academic teaching hospital were examined daily for serial features (SF), for example, wound characteristics and vital signs, in addition to standard BF, for example, wound class. Using supervised machine learning, we trained 3 Naïve Bayes classifiers (BF, SF, and BF+SF) using patient data from 1 to 5 days before diagnosis to classify SSI on the following day. For comparison, we also created a simplified SF model that used logistic regression. Control patients without SSI were matched on 5 similar consecutive postoperative days to avoid confounding by length of stay. Accuracy, sensitivity/specificity, and area under the receiver operating characteristic curve were calculated on a training and hold-out testing set.

RESULTS

Of 851 patients, 19.4% had inpatient SSIs. Univariate analysis showed differences in C-reactive protein, surgery duration, and contamination, but no differences in American Society of Anesthesiologists scores, diabetes, or emergency surgery. The BF, SF, and BF+SF classifiers had area under the receiver operating characteristic curves of 0.67, 0.76, and 0.76, respectively. The best-performing classifier (SF) had optimal sensitivity of 0.80, specificity of 0.64, positive predictive value of 0.35, and negative predictive value of 0.93. Features most associated with subsequent SSI diagnosis were granulation degree, exudate amount, nasogastric tube presence, and heart rate.

CONCLUSIONS

Serial features provided moderate positive predictive value and high negative predictive value for early identification of SSI. Addition of baseline risk factors did not improve identification. Features of evolving wound infection are discernable before the day of diagnosis, based primarily on visual inspection.

Short Term Survival after Admission for Heart Failure in Sweden: Applying Multilevel Analyses of Discriminatory Accuracy to Evaluate Institutional Performance

BACKGROUND

Hospital performance is frequently evaluated by analyzing differences between hospital averages in some quality indicators. The results are often expressed as quality charts of hospital variance (e.g., league tables, funnel plots). However, those analyses seldom consider patients heterogeneity around averages, which is of fundamental relevance for a correct evaluation. Therefore, we apply an innovative methodology based on measures of components of variance and discriminatory accuracy to analyze 30-day mortality after hospital discharge with a diagnosis of Heart Failure (HF) in Sweden.

METHODS

We analyzed 36,943 patients aged 45-80 treated in 565 wards at 71 hospitals during 2007-2009. We applied single and multilevel logistic regression analyses to calculate the odds ratios and the area under the receiver-operating characteristic (AUC). We evaluated general hospital and ward effects by quantifying the intra-class correlation coefficient (ICC) and the increment in the AUC obtained by adding random effects in a multilevel regression analysis (MLRA). Finally, the Odds Ratios (ORs) for specific ward and hospital characteristics were interpreted jointly with the proportional change in variance (PCV) and the proportion of ORs in the opposite direction (POOR).

FINDINGS

Overall, the average 30-day mortality was 9%. Using only patient information on age and previous hospitalizations for different diseases we obtained an AUC = 0.727. This value was almost unchanged when adding sex, country of birth as well as hospitals and wards levels. Average mortality was higher in small wards and municipal hospitals but the POOR values were 15% and 16% respectively.

CONCLUSIONS

Swedish wards and hospitals in general performed homogeneously well, resulting in a low 30-day mortality rate after HF. In our study, knowledge on a patient's previous hospitalizations was the best predictor of 30-day mortality, and this information did not improve by knowing the sex and country of birth of the patient or where the patient was treated.

Risk factors for surgical site infection after posterior fixation surgery and intraoperative radiotherapy for spinal metastases

PURPOSE

Posterior surgery with intraoperative radiotherapy for spinal metastases offers effective therapy, as we have reported previously. However, the procedure involves transfer from the operating room to the radiotherapy room, and as these patients are somewhat immunocompromised, the risk of postoperative surgical site infection (SSI) may be increased. The aim of our study was to identify risk factors and patient characteristics associated with postoperative SSI following posterior fixation surgery and intraoperative radiotherapy for spinal metastases.

METHODS

Participants comprised 279 patients who underwent IORT for the treatment of spinal metastases between August 2004 and June 2013. Patients who suffered SSI within 1 month after surgery were categorized as infected, and all others were categorized as non-infected. We compared factors of age, sex, use of pre-operative corticosteroid, medical history of diabetes, prognosis scores (Tomita, Tokuhashi, and Katagiri), pre- and postoperative Frankel scale scores, site of tumor origin, administration of pre-operative radiotherapy, operation time, intraoperative blood loss, intraoperative irradiation dose, and pre- and postoperative performance status between groups.

RESULTS

SSI occurred in 41 patients (14.7%). Katagiri's and Tokuhashi's prognostic scores (P < 0.05 each), postoperative Frankel scale score (P < 0.01), administration of pre-operative radiotherapy (P < 0.05), and postoperative performance status (P < 0.05) all correlated significantly with occurrence of SSI. Multivariate analysis using those factors revealed administration of pre-operative radiotherapy as a factor independently associated with SSI (P < 0.05).

CONCLUSIONS

Patient prognosis, postoperative ambulatory function, and pre-operative radiotherapy were risk factors for SSI in patients with spinal metastases. Duration of surgery and intraoperative blood loss were not associated with occurrence of SSI.