Wilton A. van Klei

Incidence of intraoperative hypotension as a function of the chosen definition: literature definitions applied to a retrospective cohort using automated data collection.

Background: Intraoperative hypotension (IOH) is a common side effect of general anesthesia and has been reported to be associated with adverse perioperative outcomes. These associations were found using different definitions for IOH. It is unknown whether the incidences of IOH found with those different definitions are comparable. The authors aimed to describe the relation between the chosen definition and incidence of IOH. Methods: First, a systematic literature search was performed to identify recent definitions of IOH that have been used in the anesthesia literature. Subsequently, these definitions were applied to a cohort of 15,509 consecutive adult patients undergoing noncardiac surgery during general anesthesia. The incidence of IOH according to the different threshold values was calculated, and the effect of a defined minimal duration of a hypotensive episode was studied. Results: Many different definitions of IOH were found. When applied to a cohort of patients, these different definitions resulted in different IOH incidences. Any episode of systolic blood pressure below 80 mmHg was found in 41% of the patients, whereas 93% of the patients had at least one episode of systolic blood pressure more than 20% below baseline. Both definitions are frequently used in the literature. The relation between threshold values from the literature and IOH incidence shows an S-shaped cumulative incidence curve, with occurrence frequencies of IOH varying from 5% to 99%. Conclusions: There is no widely accepted definition of IOH. With varying definitions, many different incidences can be reproduced. This might have implications for previously described associations between IOH and adverse outcomes.

The effect of outpatient preoperative evaluation of hospital inpatients on cancellation of surgery and length of hospital stay.

To evaluate the possible effects of outpatient preoperative evaluation (OPE) for new surgical patients who will be inpatients, we conducted an observational study at a university hospital in The Netherlands. Various outcomes before and after the introduction of an OPE clinic were compared. The study population comprised all 21,553 elective adult inpatients operated on between January 1, 1997 and December 31, 1999. Cardiac surgery, obstetric and pediatric patients, and patients operated on in same-day surgery were excluded. The main outcome measures were surgical cases canceled for medical reasons, rate of same-day admissions (who were expected to increase), and length of hospital stay. After introduction of OPE, the rate of cancellations for medical reasons decreased from 2.0% to 0.9% (adjusted odds ratio 0.7, 95% CI, 0.5–0.9). The rate of same-day admissions increased from 5.3% before to 7.7% after OPE introduction (adjusted odds ratio 1.2, 95% CI, 1.01–1.39), and the total hospital length of stay (in days) significantly decreased by a factor of 0.92 (0.90–0.94), which was partly the result of a reduction in preoperative admission time. We concluded that, although smaller than anticipated, the use of OPE for potential inpatients leads to a significant reduction of cancelled cases and of length of admission. Further increase of these benefits from OPE requires changes in institutional policy, such as forcing surgical departments to increase their number of same-day admissions.

Intraoperative Hypotension and 1-Year Mortality after Noncardiac Surgery

Background:Intraoperative hypotension (IOH) is frequently associated with adverse outcome such as 1-yr mortality. However, there is no consensus on the correct definition of IOH. The authors studied a number of different definitions of IOH, based on blood pressure thresholds and minimal episode durations, and their association with 1-yr mortality after noncardiac surgery. Methods:This cohort study included 1,705 consecutive adult patients who underwent general and vascular surgery. Data on IOH and potentially confounding variables were obtained from electronic record-keeping systems. Mortality data were collected up to 1 yr after surgery. The authors used two different techniques to reduce the influence of confounding variables, multivariable Cox proportional hazard regression modeling and classification and regression tree analysis. Results:The mortality within 1 yr after surgery was 5.2% (88 patients). After adjustment for confounding, the Cox regression analysis did not show an association between IOH and the risk of dying within 1 yr after surgery (hazard ratio around 1.00 with high P values for different definitions of IOH). Additional classification and regression tree analysis identified IOH as a predictor for 1-yr mortality in elderly patients. When the blood pressure threshold for IOH was decreased, the duration of IOH at which this association was found was decreased as well. Conclusions:This observational study showed no causal relation between IOH and 1-yr mortality after noncardiac surgery for any of the definitions of IOH. Nevertheless, additional analysis suggested that for elderly patients, the mortality risk increases when the duration of IOH becomes long enough. The length of this duration depends on the designated blood pressure threshold, suggesting that lower blood pressures are tolerated for shorter durations. The effect of IOH on 1-yr mortality remains debatable, and no firm conclusions on the lowest acceptable intraoperative blood pressures can be drawn from this study.

The Value of Routine Preoperative Electrocardiography in Predicting Myocardial Infarction After Noncardiac Surgery

Objective:The added value of a preoperative electrocardiogram (ECG) in the prediction of postoperative myocardial infarction (POMI) and death was compared with clinical risk factors identified from the patients history. Summary of Background Data:An ECG is frequently performed before surgery to screen for asymptomatic coronary artery disease. However, the value of ECG abnormalities to predict POMI has been questioned. Methods:The study included 2967 noncardiac surgery patients >50 years of age from 2 university hospitals, who were expected to stay in the hospital for >24 hours. All data were obtained from electronic record-keeping systems. Patient history and ECG abnormalities were considered as potential predictors. Multivariate logistic regression analysis was used to obtain the independent predictors of POMI and all-cause in-hospital mortality. The area under the receiver operating characteristic curve (ROC area) was estimated to evaluate the ability of different models to discriminate between patients with and without the outcome. Results:A preoperative ECG was available in 2422 patients (80%) and 1087 (45%) of the ECGs showed at least one abnormality. The ROC area of the model that included the independent predictors of POMI obtained from patient history, ie, ischemic heart disease and high-risk surgery, was 0.80. ECG abnormalities that were associated with POMI were a right and a left bundle branch block. After adding these abnormalities in the regression model, the ROC area remained 0.80. Similar results were found for all-cause mortality. Conclusions:Bundle branch blocks identified on the preoperative ECG were related to POMI and death but did not improve prediction beyond risk factors identified on patient history.

Association between intraoperative hypotension and myocardial injury after vascular surgery

Background:Postoperative myocardial injury occurs frequently after noncardiac surgery and is strongly associated with mortality. Intraoperative hypotension (IOH) is hypothesized to be a possible cause. The aim of this study was to determine the association between IOH and postoperative myocardial injury. Methods:This cohort study included 890 consecutive patients aged 60 yr or older undergoing vascular surgery from two university centers. The occurrence of myocardial injury was assessed by troponin measurements as part of a postoperative care protocol. IOH was defined by four different thresholds using either relative or absolute values of the mean arterial blood pressure based on previous studies. Either invasive or noninvasive blood pressure measurements were used. Poisson regression analysis was used to determine the association between IOH and postoperative myocardial injury, adjusted for potential clinical confounders and multiple comparisons. Results:Depending on the definition used, IOH occurred in 12 to 81% of the patients. Postoperative myocardial injury occurred in 131 (29%) patients with IOH as defined by a mean arterial pressure less than 60 mmHg, compared with 87 (20%) patients without IOH (P = 0.001). After adjustment for potential confounding factors including mean heart rates, a 40% decrease from the preinduction mean arterial blood pressure with a cumulative duration of more than 30 min was associated with postoperative myocardial injury (relative risk, 1.8; 99% CI, 1.2 to 2.6, P < 0.001). Shorter cumulative durations (less than 30 min) were not associated with myocardial injury. Postoperative myocardial infarction and death within 30 days occurred in 26 (6%) and 17 (4%) patients with IOH as defined by a mean arterial pressure less than 60 mmHg, compared with 12 (3%; P = 0.08) and 15 (3%; P = 0.77) patients without IOH, respectively. Conclusions:In elderly vascular surgery patients, IOH defined as a 40% decrease from the preinduction mean arterial blood pressure with a cumulative duration of more than 30 min was associated with postoperative myocardial injury.

Effect of β-blocker Prescription on the Incidence of Postoperative Myocardial Infarction after Hip and Knee Arthroplasty

Background:American College of Cardiology/American Heart Association guidelines recommend &bgr;-blockade for selected low- and intermediate-risk noncardiac surgery patients. The authors evaluated the effect of perioperative &bgr;-blockade on postoperative myocardial infarction (POMI) in low-risk patients undergoing intermediate-risk surgery. Methods:Patients who underwent elective hip or knee arthroplasty between January 1, 2002 and June 30, 2006 were identified. POMI was defined as a Troponin T value of more than 0.1 ng · ml−1. Patients were divided into three groups: those prescribed a &bgr;-blocker on the day of surgery and throughout their hospital stay (or 7 days, whichever came first), those prescribed a &bgr;-blocker on the day of surgery but discontinued during the first 7 days, and those not prescribed a &bgr;-blocker on the day of surgery. Propensity analysis and logistic regression were used to determine the independent association of &bgr;-blocker exposure on POMI. Results:Of the 5,158 arthroplasty patients, 992 (18%) were treated with &bgr;-blockers on the day of surgery. This &bgr;-blocker was discontinued in 252 patients (25%). POMI occurred in 77 patients (1.5%). Discontinuation of &bgr;-blocker prescription was significantly associated with POMI (odds ratio 2.0; 95% CI 1.1–3.9) and death (odds ratio 2.0; 95% CI 1.0–3.9). Conclusion:After adjustment for confounders, discontinuation of &bgr;-blocker prescription during the first week after surgery was significantly associated with POMI and death. These findings confirm the American College of Cardiology/American Heart Association Guidelines on Perioperative Cardiovascular Evaluation and Care for Noncardiac Surgery, which recommend not to withdraw &bgr;-blocker therapy.

Prediction models for clustered data: comparison of a random intercept and standard regression model

BackgroundWhen study data are clustered, standard regression analysis is considered inappropriate and analytical techniques for clustered data need to be used. For prediction research in which the interest of predictor effects is on the patient level, random effect regression models are probably preferred over standard regression analysis. It is well known that the random effect parameter estimates and the standard logistic regression parameter estimates are different. Here, we compared random effect and standard logistic regression models for their ability to provide accurate predictions.MethodsUsing an empirical study on 1642 surgical patients at risk of postoperative nausea and vomiting, who were treated by one of 19 anesthesiologists (clusters), we developed prognostic models either with standard or random intercept logistic regression. External validity of these models was assessed in new patients from other anesthesiologists. We supported our results with simulation studies using intra-class correlation coefficients (ICC) of 5%, 15%, or 30%. Standard performance measures and measures adapted for the clustered data structure were estimated.ResultsThe model developed with random effect analysis showed better discrimination than the standard approach, if the cluster effects were used for risk prediction (standard c-index of 0.69 versus 0.66). In the external validation set, both models showed similar discrimination (standard c-index 0.68 versus 0.67). The simulation study confirmed these results. For datasets with a high ICC (≥15%), model calibration was only adequate in external subjects, if the used performance measure assumed the same data structure as the model development method: standard calibration measures showed good calibration for the standard developed model, calibration measures adapting the clustered data structure showed good calibration for the prediction model with random intercept.ConclusionThe models with random intercept discriminate better than the standard model only if the cluster effect is used for predictions. The prediction model with random intercept had good calibration within clusters.

Impact of risk assessments on prophylactic antiemetic prescription and the incidence of postoperative nausea and vomiting: a cluster-randomized trial.

Background:Clinical prediction models have been shown to have moderate sensitivity and specificity, yet their use will depend on implementation in clinical practice. The authors hypothesized that implementation of a prediction model for postoperative nausea and vomiting (PONV) would lower the PONV incidence by stimulating anesthesiologists to administer more “risk-tailored” prophylaxis to patients. Methods:A single-center, cluster-randomized trial was performed in 12,032 elective surgical patients receiving anesthesia from 79 anesthesiologists. Anesthesiologists were randomized to either exposure or nonexposure to automated risk calculations for PONV (without patient-specific recommendations on prophylactic antiemetics). Anesthesiologists who treated less than 50 enrolled patients were excluded during the analysis to avoid too small clusters, yielding 11,613 patients and 57 anesthesiologists (intervention group: 5,471 and 31; care-as-usual group: 6,142 and 26). The 24-h incidence of PONV (primary outcome) and the number of prophylactic antiemetics administered per patient were studied for risk-dependent differences between allocation groups. Results:There were no differences in PONV incidence between allocation groups (crude incidence intervention group 41%, care-as-usual group 43%; odds ratio, 0.97; 95% CI, 0.87–1.1; risk-dependent odds ratio, 0.92; 95% CI, 0.80–1.1). Nevertheless, intervention-group anesthesiologists administered more prophylactic antiemetics (rate ratio, 2.0; 95% CI, 1.6–2.4) and more risk-tailored than care-as-usual–group anesthesiologists (risk-dependent rate ratio, 1.6; 95% CI, 1.3–2.0). Conclusions:Implementation of a PONV prediction model did not reduce the PONV incidence despite increased antiemetic prescription in high-risk patients by anesthesiologists. Before implementing prediction models into clinical practice, implementation studies that include patient outcomes as an endpoint are needed.

Adaptation of Clinical Prediction Models for Application in Local Settings

Background. When planning to use a validated prediction model in new patients, adequate performance is not guaranteed. For example, changes in clinical practice over time or a different case mix than the original validation population may result in inaccurate risk predictions. Objective. To demonstrate how clinical information can direct updating a prediction model and development of a strategy for handling missing predictor values in clinical practice. Methods. A previously derived and validated prediction model for postoperative nausea and vomiting was updated using a data set of 1847 patients. The update consisted of 1) changing the definition of an existing predictor, 2) reestimating the regression coefficient of a predictor, and 3) adding a new predictor to the model. The updated model was then validated in a new series of 3822 patients. Furthermore, several imputation models were considered to handle real-time missing values, so that possible missing predictor values could be anticipated during actual model use. Results. Differences in clinical practice between our local population and the original derivation population guided the update strategy of the prediction model. The predictive accuracy of the updated model was better (c statistic, 0.68; calibration slope, 1.0) than the original model (c statistic, 0.62; calibration slope, 0.57). Inclusion of logistical variables in the imputation models, besides observed patient characteristics, contributed to a strategy to deal with missing predictor values at the time of risk calculation. Conclusions. Extensive knowledge of local, clinical processes provides crucial information to guide the process of adapting a prediction model to new clinical practices.

Preoperative risk factors of intraoperative hypothermia in major surgery under general anesthesia.

UNLABELLED Preoperative factors, such as age and body habitus, affect intraoperative hypothermia during general anesthesia. In a preliminary study, we developed a logistic model to retrospectively evaluate predictors of intraoperative hypothermia in patients who received major surgery. The following factors were selected to develop the model: Z = -15.014 + 0.097 x (Age) + 0.263 x (Height) - 0.323 x (Weight) - 0.055 x (Preoperative systolic blood pressure) - 0.121 x (Preoperative heart rate). By using this model, the probability of hypothermia can be estimated by applying the following formula: Probability = 1/(1 + e(-)(Z)). If an estimated probability of hypothermia was >0.5, the sensibility of prediction was 81.5% and the specificity was 83%. In the second study, the model was applied prospectively to other patients, and the validity of the logistic model was evaluated. The core temperature showed a significant decrease in patients with a probability >0.7, who were predicted to be hypothermic, and their thermoregulatory vasoconstriction threshold also showed a significant decrease, compared with the patients with a probability <==0.3, who were predicted to be normothermic. We concluded that intraoperative hypothermia could be predicted from preoperative characteristics such as age, height, weight, systolic blood pressure, and heart rate. IMPLICATIONS Increases in age and height and decreases in weight systolic blood pressure and heart rate are major preoperative risk factors of intraoperative hypothermia during major surgery.