The Relationship Between In-House Attending Coverage and Nighttime Extubation Following Congenital Heart Surgery* Glen J. Iannucci, MD1,2; Matthew E. Oster, MD, MPH1,2; Nikhil K. Chanani, MD1,2; Scott E. Gillespie, MS3; Courtney E. McCracken, PhD3; Kirk R. Kanter, MD4; William T. Mahle, MD1,2
Objectives: Many cardiac ICUs have instituted 24/7 attending physician in-house coverage, which theoretically may allow for more expeditious weaning from ventilation and extubation. We aimed to determine whether this staffing strategy impacts rates of nighttime extubation and duration of mechanical ventilation. Design: National data were obtained from the Virtual PICU System database for all patients admitted to the cardiac ICU following congenital heart surgery in 2011 who required postoperative mechanical ventilation. Contemporaneous data from our local institution were collected in addition to the Virtual PICU System data. The combined dataset (n = 2,429) was divided based on the type of nighttime staffing model in order to compare rates of nighttime extubation and duration of mechanical ventilation between units that used an in-house attending staffing strategy and those that employed nighttime residents, fellows, or midlevel providers only. Measurements and Main Results: Institutions that currently use 24/7 in-house attending coverage did not demonstrate statistically significant differences in rates of nighttime extubation or the duration of mechanical ventilation in comparison to units without in-house attendings. Younger patients cared for in non-in-house attending units were more likely to require reintubation. *See also p. 276. 1 Sibley Heart Center, Children’s Healthcare of Atlanta, Emory University, Atlanta, GA. 2 Department of Pediatric Cardiology, Emory University, Atlanta, GA. 3 Department of Pediatrics, Emory University, Atlanta, GA. 4 Department of Cardiothoracic Surgery, Emory University, Atlanta, GA. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/ pccmjournal). The authors have disclosed that they do not have any potential conflicts of interest. Address requests for reprints to: William T. Mahle, MD, Children’s Healthcare of Atlanta, Emory University School of Medicine, 1405 Clifton Road, NE, Atlanta, GA 30322–1062. E-mail [email protected] Copyright © 2014 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies DOI: 10.1097/PCC.0000000000000068
258
www.pccmjournal.org
Conclusions: Pediatric patients who have undergone congenital heart surgery can be safely and effectively extubated without the routine presence of an attending physician. The utilization of nighttime in-house attending coverage does not appear to have significant benefits on the rate of nighttime extubation and may not reduce the duration of mechanical ventilation in units that already use in-house residents, fellows, or other midlevel providers. (Pediatr Crit Care Med 2014; 15:258–263) Key Words: 24 hour; attending physician presence; congenital heart disease; congenital heart surgery; extubation; mechanical ventilation
C
hildren and adults admitted to ICUs during the weekend or at night have an increased risk for morbidity and mortality (1–3). It has been suggested that this association is directly related to differences in staffing patterns that occur during these times. In light of these controversial findings, there has been a great deal of interest in instituting 24/7 in-house attending level staffing in pediatric, adult, and cardiac ICUs (CICUs) (4). The added experience and expertise of an in-house attending might confer a number of benefits including a reduction in the total duration of mechanical ventilation by allowing continued weaning and extubation during the nighttime hours. In fact, a recent retrospective single-center experience was published that did demonstrate a statistically significant reduction in the duration of mechanical ventilation in a PICU following institution of in-house attending coverage (5). Given the association of prolonged mechanical ventilation following congenital heart surgery with adverse outcomes (6–9), there is obvious appeal in any strategy that results in decreased duration of mechanical ventilation. Additionally, reductions in the length of mechanical ventilation are likely to result in decreased resource utilization and shortened hospital length of stay (LOS), which has been related to long-term improvements in cognitive outcome in some studies (10). March 2014 • Volume 15 • Number 3
Quality and Safety
We therefore sought to determine whether 24/7 in-house attending staffing models are associated with 1) increased rates of nighttime extubation and 2) reduced length of postoperative mechanical ventilatory support. Our current staffing model uses in-house CICU attendings from 07:00 to 17:00 on most days. During the off hours, pediatric cardiology fellows (with a minimum of 4 yr of postgraduate education) and midlevel caregivers provide in-house coverage, with attendings available from home. We hypothesized that this model is associated with similar rates of nighttime extubation and length of ventilatory support as 24/7 in-house attending staffing models.
MATERIALS AND METHODS Nighttime Extubation To assess rates of nighttime extubation and subsequent reintubation, we reviewed records of all patients admitted to the CICU following congenital heart surgery during the calendar year of 2011. In order to obtain a national sample, we used the Virtual PICU System (VPS) database, a clinical database dedicated to standardized data sharing and benchmarking among PICUs. Contemporaneous data were collected from our local institution, Children’s Healthcare of Atlanta. In 2011, our center did not yet contribute CICU data to the VPS, so the VPS relevant data fields were extracted from an institutional database for the purpose of this analysis. In order to reduce variability, patients from the VPS dataset were only included if they were cared for in a dedicated CICU. Patients were excluded if their primary surgical procedure was extracorporeal membrane oxygenation cannulation. For patients who underwent multiple surgical procedures during the same hospitalization (i.e., shunt revision), only the data on the index surgical procedure were included. Patients who were extubated between 17:00 and 07:00 were classified as nighttime extubations. Particular attention was given to the need for and timing of reintubation. Patients who were reintubated within 48 hours of extubation were classified as failing extubation. Subgroup analysis was then performed evaluating neonates (defined as < 28 days old), infants (defined as < 1 year old), and children (all patients > 1 year old). Additionally, patients were subclassified based on surgical type and complexity using Risk Adjusted Congenital Heart Surgery (RACHS-1) criteria (11). We then evaluated the association of in-house attending coverage with nighttime extubation by comparing extubation patterns from institutions that employ in-house attending coverage in the care of postoperative patients following congenital heart surgery. The 1,794 subjects from the VPS database that had complete data regarding the precise timing of intubation and extubation were combined with our local data (n = 635) to produce the study population (n = 2,429). Duration of Mechanical Ventilation An additional analysis was then carried out to evaluate the association of in-house attending coverage with length of ventilatory support. In order to provide for more direct comparisons, surgical procedures were selected that were relatively Pediatric Critical Care Medicine
homogenous with regard to risk stratification and need for postoperative ventilation. Finally, multiple linear regression was used to determine if the presence of in-house attending coverage significantly reduced duration of mechanical ventilation while controlling for the effects of other clinical and demographic factors. Statistical Methods Descriptive statistics were calculated for all variables of interest using counts and frequencies, medians and ranges, or means and CIs. Within each cohort, patients were categorized based on the primary cardiac procedure and the time of their extubation (day vs night) or nighttime coverage (attending vs resident/fellow). Normality was assessed by visually inspecting histograms and using the Anderson-Darling test for normality. The primary outcome variables, duration of mechanical ventilation, were right skewed; therefore, the data were log-transformed prior to statistical testing. Two-sample independent t tests were used to compare the log duration of mechanical ventilation. Because the results were in the logarithmic scale, means and 95% CIs were back-transformed into the original units by exponentiation. The resulting back-transformed data represent the geometric mean of the original data (12). For some variables (age and weight), the Mann-Whitney U or the two-sample Kolmogorov-Smirnov tests were used to compare two groups because the data were nonnormally distributed and attempts to transform the data were unsuccessful. Chi-square tests were used to compare categorical variables between groups. Lastly, the effect of in-house attending coverage versus fellow or midlevel provider coverage on log duration of mechanical ventilation was modeled using multiple linear regression while adjusting for factors such as RACHS-1 score, age and weight, day versus night extubation, and hospital location. Resulting coefficients and CIs were back-transformed to the original scale. Statistical significance was assessed using a significance level of 0.05 using two-sided statistical tests. All statistical analyses were performed using SAS 9.2 (Cary, NC) and the R Statistical Computing Language (Vienna, Austria). This study was performed with the approval of our institution’s institutional review board.
RESULTS Nighttime Extubation During the 1-year study period, there were 1,794 surgical cases from the VPS dataset that met inclusion criteria. Of these cases, 47.6% had a RACHS-1 score greater than two. These patients were cared for in 14 distinct units. At our institution, there were 635 surgical cases that met inclusion criteria. Of these cases, 48.2% of the surgical procedures had a RACHS-1 score greater than two. One third (33.2%, n = 807) of all extubations occurred during the hours of 17:00 to 07:00. There were relatively consistent rates of extubation throughout the course of the night (Fig. 1). When evaluating the combined dataset in Table 1, patients who were extubated during the day were significantly more likely to require reintubation within 48 hours www.pccmjournal.org
259
Iannucci et al
Figure 1. Timing of extubation of the study population.
than those patients who were extubated at night (7.4% vs 4.1%, p = 0.002). Additionally, compared to patients extubated during the day, patients extubated at night were significantly older (212.4 d vs 72.8 d; p < 0.001) and had lower RACHS-1 scores (p < 0.001) (Table 1). As one would expect, there was a significantly higher rate of reintubation for patients who were younger. In fact, patients who were less than 28 days old at the time of their surgical procedure required reintubation within 48 hours of extubation 8.4% of the time. This is in contrast to patients who were more than 1 year old at the time of surgery who had a reintubation rate of only 3.4% (p < 0.001). Forty-six percent of surgical cases were cared for in units with a nighttime attending in-house model. These institutions did not extubate more patients during the nighttime hours than those who did not have an attending physician in-house.
In fact, the converse was true. Units that were covered by residents or fellows actually extubated more patients during the nighttime hours than units with routine in-house attending coverage (38.1% vs 27.4%; p < 0.001). In evaluating the potential role of nighttime attending presence on the rate of reintubation from Table 2, we found that patients who were cared for under a nonattending model at night had higher rates of reintubation (7.9% vs 4.4%; p < 0.001). Table 3 demonstrates that there was no statistically significant difference in the number of patients who were extubated at night (3.9% vs 4.2%; p = 0.865). Duration of Mechanical Ventilation In evaluating the groups of patients who were managed in a unit with 24/7 in-house attending coverage, and those who were not, there were statistically significant differences in the study
Table 1. Comparison of Patient Demographics for Those Extubated During the Day and Those at Night Extubation Demographic/Characteristic
Age (d), median (minimum–maximum) Weight (kg), median (minimum–maximum) Risk Adjusted Congenital Heart Surgery score ≥ 3, n (%) Chromosomal anomaly, n (%)
a
Preoperative mechanical ventilation, n (%)
a
Rates of reintubation, n (%)
Day (n = 1,622)
Night (n = 807)
p
72.8 (0–16,171.1)
212.4 (0–19,169.1)
< 0.001
4.3 (1.3–116.2)
6.9 (1.4–116.7)
< 0.001
903 (55.7)
349 (43.3)
< 0.001
64 (19.9)
42 (13.4)
0.029
30 (9.3)
16 (5.1)
0.041
120 (7.4)
33 (4.1)
0.002
Data only available from Children’s Healthcare of Atlanta (n day = 322, n night = 313).
a
260
www.pccmjournal.org
March 2014 • Volume 15 • Number 3
Quality and Safety
Table 2.
Comparison of Patient Demographics by Nighttime Physician Coverage Nighttime Coverage p
Demographic/Characteristic
Attending (n = 1,113)
Resident/Fellow (n = 1,316)
Age (d), median (minimum–maximum)
111.0 (0–19,169.1)
127.4 (0–16,980.1)
0.020
4.8 (1.4–116.7)
5.4 (1.3–102.0)
0.042
Weight (kg), median (minimum–maximum)
638 (57.3)
Risk Adjusted Congenital Heart Surgery score ≥ 3, n (%)
49 (4.4)
Rates of reintubation, n (%)
Table 3.
614 (46.7)
< 0.001
104 (7.9)
< 0.001
Rates of Reintubation Broken Down by Age and Time of Extubation Attending
Characteristic
Age
Time of extubation
Resident/Fellow Number
Reintubated (%)
p
21 (5.2)
425
49 (11.5)
< 0.001
377
18 (4.8)
463
39 (8.4)
0.037
328
10 (3.1)
428
16 (3.7)
0.603
Day
808
37 (4.6)
814
83 (10.2)
Night
305
12 (3.9)
502
21 (4.2)
Level
Number
< 28 d
408
28 d–1 yr > 1 yr
Reintubated (%)
populations (Table 2). Procedures that were homogenous with regard to their risk profile and need for postoperative ventilation were selected for evaluation (Supplemental Table 1, Supplemental Digital Content 1, http://links.lww.com/PCC/A83). Of the seven procedures considered, only ventriculoseptal defect repair presented a significant difference in the length of mechanical ventilation for in-house attending physicians versus resident/ fellow coverage (22.7 vs 15.3 hr; p = 0.033). In the model that adjusted for RACHS score, weight, extubation time, reintubation, and hospital, presence of an in-house attending physician failed to significantly impact the duration of mechanical ventilation.
DISCUSSION In this study, we found that institutions that use fellows or midlevel providers can achieve high rates of successful nighttime extubation, in appropriately selected, low-risk patients, with a low risk of reintubation. Furthermore, in evaluating national data through the use of the VPS database, it becomes clear that CICUs that do use a nighttime attending in-house staffing model do not extubate patients more frequently at night than those units that do not. Additionally, the presence of 24/7 in-house attending coverage does not appear to significantly impact the duration of mechanical ventilation. Recently, the American College of Critical Care Medicine recommended the adoption of 24/7 in-house attending physician coverage of ICUs. This recommendation was made on the basis of adult data (1, 2, 4, 13–18). The presence of in-house attending coverage has been prospectively addressed by a few studies in the adult literature. A single-center experience that evaluated care in an adult ICU at an academic medical center before and after conversion to 24/7 in-house attending coverage demonstrated improvement in adherence to established Pediatric Critical Care Medicine
< 0.001 0.865
medical pathways, improvements in patient and staff satisfaction, and shortened hospital LOS (19). An additional study in an adult oncology ICU demonstrated improvements in overall LOS and bed turnover rates (20). A more recent study evaluating outcomes following conversion from an open ICU staffing model to a closed model using in-house residents for night coverage did demonstrate significant improvements in multiple outcomes including duration of mechanical ventilation. It, however, remains unclear if the addition of nighttime attending coverage would result in further gains (21). In evaluating adult patients following open-heart surgery, there was an association between the transition to 24/7 in-house coverage and a reduction in the duration of mechanical ventilation; however, there were many additional factors that changed during the study period (22). With regard to pediatric units, a retrospective review of newborns admitted to a neonatal ICU in Canada revealed increased mortality in patients who were admitted during the nighttime hours (23). More recently, Nishisaki et al (5) reported their single-center experience that analyzed the effects of conversion from a 12-hour in-house attending model to a 24-hour in-house attending model in the PICU. This study did demonstrate a reduction in the duration of mechanical ventilation and speculated that this may have been due to more timely extubation. To the authors’ knowledge, no formal evaluation of the presence of nighttime attending coverage in the pediatric CICU has been undertaken. Some of the proponents for in-house attending coverage have noted that this would likely reduce delays in initiating care changes and facilitate ventilator weaning overnight (24). The potential benefit to having experienced in-house coverage in the setting of a pediatric CICU is inherent; however, published data confirming this benefit are sparse. The current www.pccmjournal.org
261
Iannucci et al
study does suggest that the presence of an in-house attending staffing model may be associated with a decreased risk of reintubation in infants who underwent surgery prior. The finding that those patients who were extubated during the daytime hours were more likely to require reintubation, irrespective of the staffing model in which they were cared for likely reflects the increased overall complexity of the patients who were extubated during the daytime hours. Prior studies have shown that prolonged duration of mechanical ventilation and the resultant increase in hospital LOS have been associated with increased morbidity and mortality (6, 9, 10, 25). Any strategy that could result in decreased duration of mechanical ventilation would likely benefit this patient population significantly. Based on the current study, it seems that in a unit that already uses experienced personnel at night, the addition of a nighttime attending physician would be of limited utility in achieving more frequent nighttime extubation. Furthermore, in reviewing the data in Table 1, it seems that in-house attending coverage plays a relatively small role in the determination of the duration of mechanical ventilation. This was investigated further and bolstered by the results from the multiple linear regression model. Numerous studies have demonstrated the safety and feasibility of rapid extubation of pediatric patients undergoing congenital heart surgery (26–33). In institutions that routinely extubate lowrisk patients in the operating room, it may be even more challenging to realize a reduction in the duration of mechanical ventilation through the addition of a nighttime attending.
LIMITATIONS Our study is limited by several assumptions that were made in the study design. In our current staffing model, CICU attending physicians take home call but are present for all admissions and routinely come in for any major complications or significant clinical deterioration. We suspect that this type of staffing is similar at many of the other institutions that do not have routine in-house coverage. As such it is conceivable that an attending may have been present during the nighttime hours when a patient was extubated. This would likely have represented a very small number of cases, and the inclusion of these cases was felt to be unlikely to impact the overall study outcome. With regard to the examination of the VPS dataset, the main limitation was the voluntary nature of data submission regarding the timing of intubation and/or extubation. As such, in spite of the multiple centers contributing data to the VPS dataset, our evaluation of extubation timing was limited to only those patients with documented intubation and extubation times; thus reducing the effective sample size. Therefore, it is possible that given our reduced sample size, we were underpowered to detect the trend toward reduced duration of mechanical ventilation in patients cared for under an in-house attending model in comparison to those patients who were cared for in units that used other staffing models. Additionally, our study specifically evaluated patients who were mechanically ventilated at the time of postoperative CICU admission. 262
www.pccmjournal.org
Thus, the least complicated patients were less likely to be included in the analysis.
CONCLUSIONS In this study, we found that there are relatively high rates of successful nighttime extubation in the absence of an in-house attending physician. Furthermore, in evaluating national data, it is noteworthy that institutions that have nighttime in-house attending presence do not extubate more patients at night than those that use more traditional staffing models, nor did they demonstrate a consistent reduction in the duration of mechanical ventilation. Although there may be benefits to the presence of an in-house attending physician in the care of patients following congenital heart surgery, the addition of 24/7 in-house attending coverage in a CICU that already uses in-house residents, fellows, or midlevel providers may not significantly reduce the duration of mechanical ventilation.
ACKNOWLEDGMENTS Virtual PICU System (VPS) data were provided from the VPS. No endorsement or editorial restriction of the interpretation of these data or opinions of the authors has been implied or stated. The authors acknowledge Pam Carlock, RN, our surgical database manager, for assistance in accessing the surgical database at Children’s Healthcare of Atlanta.
REFERENCES
1. Cavallazzi R, Marik PE, Hirani A, et al: Association between time of admission to the ICU and mortality: A systematic review and metaanalysis. Chest 2010; 138:68–75 2. Angus DC, Dorman T, Robinson KA, et al: Physician staffing patterns and clinical outcomes in critically ill patients: A systematic review. JAMA 2002; 288:2151–2162 3. Lee SK, Lee DS, Andrews WL, et al: Higher mortality rates among inborn infants admitted to neonatal intensive care units at night. J Pediatr 2003; 143:592–597 4. Wallace DJ, Angus DC, Barnato AE, et al: Nighttime intensivist staffing and mortality among critically ill patients. N Engl J Med 2012; 366:2093–2101 5. Nishisaki A, Pines JM, Lin R, et al: The impact of 24-hr, in-hospital pediatric critical care attending physician presence on process of care and patient outcomes. Crit Care Med 2012; 40:2190–2195 6. Roeleveld PP, Guijt D, Kuijper EJ, et al: Ventilator-associated pneumonia in children after cardiac surgery in The Netherlands. Intensive Care Med 2011; 37:1656–1663 7. Shi S, Zhao Z, Liu X, et al: Perioperative risk factors for prolonged mechanical ventilation following cardiac surgery in neonates and young infants. Chest 2008; 134:768–774 8. Polito A, Patorno E, Costello JM, et al: Perioperative factors associated with prolonged mechanical ventilation after complex congenital heart surgery. Pediatr Crit Care Med 2011; 12:e122–e126 9. Costello JM, Graham DA, Morrow DF, et al: Risk factors for central line-associated bloodstream infection in a pediatric cardiac intensive care unit. Pediatr Crit Care Med 2009; 10:453–459 10. Newburger JW, Wypij D, Bellinger DC, et al: Length of stay after infant heart surgery is related to cognitive outcome at age 8 years. J Pediatr 2003; 143:67–73 11. Jenkins KJ, Gauvreau K, Newburger JW, et al: Consensus-based method for risk adjustment for surgery for congenital heart disease. J Thorac Cardiovasc Surg 2002; 123:110–118 March 2014 • Volume 15 • Number 3
Quality and Safety 12. Bland JM, Altman DG: Transformations, means, and confidence intervals. BMJ 1996; 312:1079 13. Uusaro A, Kari A, Ruokonen E: The effects of ICU admission and discharge times on mortality in Finland. Intensive Care Med 2003; 29:2144–2148 14. Ensminger SA, Morales IJ, Peters SG, et al: The hospital mortality of patients admitted to the ICU on weekends. Chest 2004; 126:1292–1298 15. Luyt CE, Combes A, Aegerter P, et al: Mortality among patients admitted to intensive care units during weekday day shifts compared with “off” hours. Crit Care Med 2007; 35:3–11 16. Gasperino J: The Leapfrog initiative for intensive care unit physician staffing and its impact on intensive care unit performance: A narrative review. Health Policy 2011; 102:223–228 17. Manthous CA: Leapfrog and critical care: Evidence- and reality-based intensive care for the 21st century. Am J Med 2004; 116:188–193 18. Haupt MT, Bekes CE, Brilli RJ, et al; Task Force of the American College of Critical Care Medicine, Society of Critical Care Medicine: Guidelines on critical care services and personnel: Recommendations based on a system of categorization of three levels of care. Crit Care Med 2003; 31:2677–2683 19. Gajic O, Afessa B, Hanson AC, et al: Effect of 24-hour mandatory versus on-demand critical care specialist presence on quality of care and family and provider satisfaction in the intensive care unit of a teaching hospital. Crit Care Med 2008; 36:36–44 20. Hawari FI, Al Najjar TI, Zaru L, et al: The effect of implementing high-intensity intensive care unit staffing model on outcome of critically ill oncology patients. Crit Care Med 2009; 37:1967–1971 21. Parikh A, Huang SA, Murthy P, et al: Quality improvement and cost savings after implementation of the Leapfrog intensive care unit physician staffing standard at a community teaching hospital. Crit Care Med 2012; 40:2754–2759 22. Kumar K, Zarychanski R, Bell DD, et al; Cardiovascular Health Research in Manitoba Investigator Group: Impact of 24-hour in-house
Pediatric Critical Care Medicine
intensivists on a dedicated cardiac surgery intensive care unit. Ann Thorac Surg 2009; 88:1153–1161 23. Lee SK, Lee DS, Andrews WL, et al; Canadian Neonatal Network: Higher mortality rates among inborn infants admitted to neonatal intensive care units at night. J Pediatr 2003; 143:592–597 24. Farnan JM, Johnson JK, Meltzer DO, et al: Resident uncertainty in clinical decision making and impact on patient care: A qualitative study. Qual Saf Health Care 2008; 17:122–126 25. Wernovsky G, Kuijpers M, Van Rossem MC, et al: Postoperative course in the cardiac intensive care unit following the first stage of Norwood reconstruction. Cardiol Young 2007; 17:652–665 26. Cooper DS, Costello JM, Bronicki RA, et al: Current challenges in cardiac intensive care: Optimal strategies for mechanical ventilation and timing of extubation. Cardiol Young 2008; 18(Suppl 3):72–83 27. Karthekeyan RB, Sundar AS, Sulaiman S, et al: Early extubation in tetralogy of Fallot patients after complete repair. Cardiol Young 2011; 21:378–382 28. Winch PD, Nicholson L, Isaacs J, et al: Predictors of successful early extubation following congenital cardiac surgery in neonates and infants. Heart Lung Circ 2009; 18:271–276 29. Yamasaki Y, Shime N, Miyazaki T, et al: Fast-track postoperative care for neonatal cardiac surgery: A single-institute experience. J Anesth 2011; 25:321–329 30. Preisman S, Lembersky H, Yusim Y, et al: A randomized trial of outcomes of anesthetic management directed to very early extubation after cardiac surgery in children. J Cardiothorac Vasc Anesth 2009; 23:348–357 31. Davis S, Worley S, Mee RB, et al: Factors associated with early extubation after cardiac surgery in young children. Pediatr Crit Care Med 2004; 5:63–68 32. Vricella LA, Dearani JA, Gundry SR, et al: Ultra fast track in elective congenital cardiac surgery. Ann Thorac Surg 2000; 69:865–871 33. Mittnacht AJ, Thanjan M, Srivastava S, et al: Extubation in the operating room after congenital heart surgery in children. J Thorac Cardiovasc Surg 2008; 136:88–93
www.pccmjournal.org
263
Objectives: Many cardiac ICUs have instituted 24/7 attending physician in-house coverage, which theoretically may allow for more expeditious weaning from ventilation and extubation. We aimed to determine whether this staffing strategy impacts rates of nighttime extubation and duration of mechanical ventilation. Design: National data were obtained from the Virtual PICU System database for all patients admitted to the cardiac ICU following congenital heart surgery in 2011 who required postoperative mechanical ventilation. Contemporaneous data from our local institution were collected in addition to the Virtual PICU System data. The combined dataset (n = 2,429) was divided based on the type of nighttime staffing model in order to compare rates of nighttime extubation and duration of mechanical ventilation between units that used an in-house attending staffing strategy and those that employed nighttime residents, fellows, or midlevel providers only. Measurements and Main Results: Institutions that currently use 24/7 in-house attending coverage did not demonstrate statistically significant differences in rates of nighttime extubation or the duration of mechanical ventilation in comparison to units without in-house attendings. Younger patients cared for in non-in-house attending units were more likely to require reintubation. *See also p. 276. 1 Sibley Heart Center, Children’s Healthcare of Atlanta, Emory University, Atlanta, GA. 2 Department of Pediatric Cardiology, Emory University, Atlanta, GA. 3 Department of Pediatrics, Emory University, Atlanta, GA. 4 Department of Cardiothoracic Surgery, Emory University, Atlanta, GA. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/ pccmjournal). The authors have disclosed that they do not have any potential conflicts of interest. Address requests for reprints to: William T. Mahle, MD, Children’s Healthcare of Atlanta, Emory University School of Medicine, 1405 Clifton Road, NE, Atlanta, GA 30322–1062. E-mail [email protected] Copyright © 2014 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies DOI: 10.1097/PCC.0000000000000068
258
www.pccmjournal.org
Conclusions: Pediatric patients who have undergone congenital heart surgery can be safely and effectively extubated without the routine presence of an attending physician. The utilization of nighttime in-house attending coverage does not appear to have significant benefits on the rate of nighttime extubation and may not reduce the duration of mechanical ventilation in units that already use in-house residents, fellows, or other midlevel providers. (Pediatr Crit Care Med 2014; 15:258–263) Key Words: 24 hour; attending physician presence; congenital heart disease; congenital heart surgery; extubation; mechanical ventilation
C
hildren and adults admitted to ICUs during the weekend or at night have an increased risk for morbidity and mortality (1–3). It has been suggested that this association is directly related to differences in staffing patterns that occur during these times. In light of these controversial findings, there has been a great deal of interest in instituting 24/7 in-house attending level staffing in pediatric, adult, and cardiac ICUs (CICUs) (4). The added experience and expertise of an in-house attending might confer a number of benefits including a reduction in the total duration of mechanical ventilation by allowing continued weaning and extubation during the nighttime hours. In fact, a recent retrospective single-center experience was published that did demonstrate a statistically significant reduction in the duration of mechanical ventilation in a PICU following institution of in-house attending coverage (5). Given the association of prolonged mechanical ventilation following congenital heart surgery with adverse outcomes (6–9), there is obvious appeal in any strategy that results in decreased duration of mechanical ventilation. Additionally, reductions in the length of mechanical ventilation are likely to result in decreased resource utilization and shortened hospital length of stay (LOS), which has been related to long-term improvements in cognitive outcome in some studies (10). March 2014 • Volume 15 • Number 3
Quality and Safety
We therefore sought to determine whether 24/7 in-house attending staffing models are associated with 1) increased rates of nighttime extubation and 2) reduced length of postoperative mechanical ventilatory support. Our current staffing model uses in-house CICU attendings from 07:00 to 17:00 on most days. During the off hours, pediatric cardiology fellows (with a minimum of 4 yr of postgraduate education) and midlevel caregivers provide in-house coverage, with attendings available from home. We hypothesized that this model is associated with similar rates of nighttime extubation and length of ventilatory support as 24/7 in-house attending staffing models.
MATERIALS AND METHODS Nighttime Extubation To assess rates of nighttime extubation and subsequent reintubation, we reviewed records of all patients admitted to the CICU following congenital heart surgery during the calendar year of 2011. In order to obtain a national sample, we used the Virtual PICU System (VPS) database, a clinical database dedicated to standardized data sharing and benchmarking among PICUs. Contemporaneous data were collected from our local institution, Children’s Healthcare of Atlanta. In 2011, our center did not yet contribute CICU data to the VPS, so the VPS relevant data fields were extracted from an institutional database for the purpose of this analysis. In order to reduce variability, patients from the VPS dataset were only included if they were cared for in a dedicated CICU. Patients were excluded if their primary surgical procedure was extracorporeal membrane oxygenation cannulation. For patients who underwent multiple surgical procedures during the same hospitalization (i.e., shunt revision), only the data on the index surgical procedure were included. Patients who were extubated between 17:00 and 07:00 were classified as nighttime extubations. Particular attention was given to the need for and timing of reintubation. Patients who were reintubated within 48 hours of extubation were classified as failing extubation. Subgroup analysis was then performed evaluating neonates (defined as < 28 days old), infants (defined as < 1 year old), and children (all patients > 1 year old). Additionally, patients were subclassified based on surgical type and complexity using Risk Adjusted Congenital Heart Surgery (RACHS-1) criteria (11). We then evaluated the association of in-house attending coverage with nighttime extubation by comparing extubation patterns from institutions that employ in-house attending coverage in the care of postoperative patients following congenital heart surgery. The 1,794 subjects from the VPS database that had complete data regarding the precise timing of intubation and extubation were combined with our local data (n = 635) to produce the study population (n = 2,429). Duration of Mechanical Ventilation An additional analysis was then carried out to evaluate the association of in-house attending coverage with length of ventilatory support. In order to provide for more direct comparisons, surgical procedures were selected that were relatively Pediatric Critical Care Medicine
homogenous with regard to risk stratification and need for postoperative ventilation. Finally, multiple linear regression was used to determine if the presence of in-house attending coverage significantly reduced duration of mechanical ventilation while controlling for the effects of other clinical and demographic factors. Statistical Methods Descriptive statistics were calculated for all variables of interest using counts and frequencies, medians and ranges, or means and CIs. Within each cohort, patients were categorized based on the primary cardiac procedure and the time of their extubation (day vs night) or nighttime coverage (attending vs resident/fellow). Normality was assessed by visually inspecting histograms and using the Anderson-Darling test for normality. The primary outcome variables, duration of mechanical ventilation, were right skewed; therefore, the data were log-transformed prior to statistical testing. Two-sample independent t tests were used to compare the log duration of mechanical ventilation. Because the results were in the logarithmic scale, means and 95% CIs were back-transformed into the original units by exponentiation. The resulting back-transformed data represent the geometric mean of the original data (12). For some variables (age and weight), the Mann-Whitney U or the two-sample Kolmogorov-Smirnov tests were used to compare two groups because the data were nonnormally distributed and attempts to transform the data were unsuccessful. Chi-square tests were used to compare categorical variables between groups. Lastly, the effect of in-house attending coverage versus fellow or midlevel provider coverage on log duration of mechanical ventilation was modeled using multiple linear regression while adjusting for factors such as RACHS-1 score, age and weight, day versus night extubation, and hospital location. Resulting coefficients and CIs were back-transformed to the original scale. Statistical significance was assessed using a significance level of 0.05 using two-sided statistical tests. All statistical analyses were performed using SAS 9.2 (Cary, NC) and the R Statistical Computing Language (Vienna, Austria). This study was performed with the approval of our institution’s institutional review board.
RESULTS Nighttime Extubation During the 1-year study period, there were 1,794 surgical cases from the VPS dataset that met inclusion criteria. Of these cases, 47.6% had a RACHS-1 score greater than two. These patients were cared for in 14 distinct units. At our institution, there were 635 surgical cases that met inclusion criteria. Of these cases, 48.2% of the surgical procedures had a RACHS-1 score greater than two. One third (33.2%, n = 807) of all extubations occurred during the hours of 17:00 to 07:00. There were relatively consistent rates of extubation throughout the course of the night (Fig. 1). When evaluating the combined dataset in Table 1, patients who were extubated during the day were significantly more likely to require reintubation within 48 hours www.pccmjournal.org
259
Iannucci et al
Figure 1. Timing of extubation of the study population.
than those patients who were extubated at night (7.4% vs 4.1%, p = 0.002). Additionally, compared to patients extubated during the day, patients extubated at night were significantly older (212.4 d vs 72.8 d; p < 0.001) and had lower RACHS-1 scores (p < 0.001) (Table 1). As one would expect, there was a significantly higher rate of reintubation for patients who were younger. In fact, patients who were less than 28 days old at the time of their surgical procedure required reintubation within 48 hours of extubation 8.4% of the time. This is in contrast to patients who were more than 1 year old at the time of surgery who had a reintubation rate of only 3.4% (p < 0.001). Forty-six percent of surgical cases were cared for in units with a nighttime attending in-house model. These institutions did not extubate more patients during the nighttime hours than those who did not have an attending physician in-house.
In fact, the converse was true. Units that were covered by residents or fellows actually extubated more patients during the nighttime hours than units with routine in-house attending coverage (38.1% vs 27.4%; p < 0.001). In evaluating the potential role of nighttime attending presence on the rate of reintubation from Table 2, we found that patients who were cared for under a nonattending model at night had higher rates of reintubation (7.9% vs 4.4%; p < 0.001). Table 3 demonstrates that there was no statistically significant difference in the number of patients who were extubated at night (3.9% vs 4.2%; p = 0.865). Duration of Mechanical Ventilation In evaluating the groups of patients who were managed in a unit with 24/7 in-house attending coverage, and those who were not, there were statistically significant differences in the study
Table 1. Comparison of Patient Demographics for Those Extubated During the Day and Those at Night Extubation Demographic/Characteristic
Age (d), median (minimum–maximum) Weight (kg), median (minimum–maximum) Risk Adjusted Congenital Heart Surgery score ≥ 3, n (%) Chromosomal anomaly, n (%)
a
Preoperative mechanical ventilation, n (%)
a
Rates of reintubation, n (%)
Day (n = 1,622)
Night (n = 807)
p
72.8 (0–16,171.1)
212.4 (0–19,169.1)
< 0.001
4.3 (1.3–116.2)
6.9 (1.4–116.7)
< 0.001
903 (55.7)
349 (43.3)
< 0.001
64 (19.9)
42 (13.4)
0.029
30 (9.3)
16 (5.1)
0.041
120 (7.4)
33 (4.1)
0.002
Data only available from Children’s Healthcare of Atlanta (n day = 322, n night = 313).
a
260
www.pccmjournal.org
March 2014 • Volume 15 • Number 3
Quality and Safety
Table 2.
Comparison of Patient Demographics by Nighttime Physician Coverage Nighttime Coverage p
Demographic/Characteristic
Attending (n = 1,113)
Resident/Fellow (n = 1,316)
Age (d), median (minimum–maximum)
111.0 (0–19,169.1)
127.4 (0–16,980.1)
0.020
4.8 (1.4–116.7)
5.4 (1.3–102.0)
0.042
Weight (kg), median (minimum–maximum)
638 (57.3)
Risk Adjusted Congenital Heart Surgery score ≥ 3, n (%)
49 (4.4)
Rates of reintubation, n (%)
Table 3.
614 (46.7)
< 0.001
104 (7.9)
< 0.001
Rates of Reintubation Broken Down by Age and Time of Extubation Attending
Characteristic
Age
Time of extubation
Resident/Fellow Number
Reintubated (%)
p
21 (5.2)
425
49 (11.5)
< 0.001
377
18 (4.8)
463
39 (8.4)
0.037
328
10 (3.1)
428
16 (3.7)
0.603
Day
808
37 (4.6)
814
83 (10.2)
Night
305
12 (3.9)
502
21 (4.2)
Level
Number
< 28 d
408
28 d–1 yr > 1 yr
Reintubated (%)
populations (Table 2). Procedures that were homogenous with regard to their risk profile and need for postoperative ventilation were selected for evaluation (Supplemental Table 1, Supplemental Digital Content 1, http://links.lww.com/PCC/A83). Of the seven procedures considered, only ventriculoseptal defect repair presented a significant difference in the length of mechanical ventilation for in-house attending physicians versus resident/ fellow coverage (22.7 vs 15.3 hr; p = 0.033). In the model that adjusted for RACHS score, weight, extubation time, reintubation, and hospital, presence of an in-house attending physician failed to significantly impact the duration of mechanical ventilation.
DISCUSSION In this study, we found that institutions that use fellows or midlevel providers can achieve high rates of successful nighttime extubation, in appropriately selected, low-risk patients, with a low risk of reintubation. Furthermore, in evaluating national data through the use of the VPS database, it becomes clear that CICUs that do use a nighttime attending in-house staffing model do not extubate patients more frequently at night than those units that do not. Additionally, the presence of 24/7 in-house attending coverage does not appear to significantly impact the duration of mechanical ventilation. Recently, the American College of Critical Care Medicine recommended the adoption of 24/7 in-house attending physician coverage of ICUs. This recommendation was made on the basis of adult data (1, 2, 4, 13–18). The presence of in-house attending coverage has been prospectively addressed by a few studies in the adult literature. A single-center experience that evaluated care in an adult ICU at an academic medical center before and after conversion to 24/7 in-house attending coverage demonstrated improvement in adherence to established Pediatric Critical Care Medicine
< 0.001 0.865
medical pathways, improvements in patient and staff satisfaction, and shortened hospital LOS (19). An additional study in an adult oncology ICU demonstrated improvements in overall LOS and bed turnover rates (20). A more recent study evaluating outcomes following conversion from an open ICU staffing model to a closed model using in-house residents for night coverage did demonstrate significant improvements in multiple outcomes including duration of mechanical ventilation. It, however, remains unclear if the addition of nighttime attending coverage would result in further gains (21). In evaluating adult patients following open-heart surgery, there was an association between the transition to 24/7 in-house coverage and a reduction in the duration of mechanical ventilation; however, there were many additional factors that changed during the study period (22). With regard to pediatric units, a retrospective review of newborns admitted to a neonatal ICU in Canada revealed increased mortality in patients who were admitted during the nighttime hours (23). More recently, Nishisaki et al (5) reported their single-center experience that analyzed the effects of conversion from a 12-hour in-house attending model to a 24-hour in-house attending model in the PICU. This study did demonstrate a reduction in the duration of mechanical ventilation and speculated that this may have been due to more timely extubation. To the authors’ knowledge, no formal evaluation of the presence of nighttime attending coverage in the pediatric CICU has been undertaken. Some of the proponents for in-house attending coverage have noted that this would likely reduce delays in initiating care changes and facilitate ventilator weaning overnight (24). The potential benefit to having experienced in-house coverage in the setting of a pediatric CICU is inherent; however, published data confirming this benefit are sparse. The current www.pccmjournal.org
261
Iannucci et al
study does suggest that the presence of an in-house attending staffing model may be associated with a decreased risk of reintubation in infants who underwent surgery prior. The finding that those patients who were extubated during the daytime hours were more likely to require reintubation, irrespective of the staffing model in which they were cared for likely reflects the increased overall complexity of the patients who were extubated during the daytime hours. Prior studies have shown that prolonged duration of mechanical ventilation and the resultant increase in hospital LOS have been associated with increased morbidity and mortality (6, 9, 10, 25). Any strategy that could result in decreased duration of mechanical ventilation would likely benefit this patient population significantly. Based on the current study, it seems that in a unit that already uses experienced personnel at night, the addition of a nighttime attending physician would be of limited utility in achieving more frequent nighttime extubation. Furthermore, in reviewing the data in Table 1, it seems that in-house attending coverage plays a relatively small role in the determination of the duration of mechanical ventilation. This was investigated further and bolstered by the results from the multiple linear regression model. Numerous studies have demonstrated the safety and feasibility of rapid extubation of pediatric patients undergoing congenital heart surgery (26–33). In institutions that routinely extubate lowrisk patients in the operating room, it may be even more challenging to realize a reduction in the duration of mechanical ventilation through the addition of a nighttime attending.
LIMITATIONS Our study is limited by several assumptions that were made in the study design. In our current staffing model, CICU attending physicians take home call but are present for all admissions and routinely come in for any major complications or significant clinical deterioration. We suspect that this type of staffing is similar at many of the other institutions that do not have routine in-house coverage. As such it is conceivable that an attending may have been present during the nighttime hours when a patient was extubated. This would likely have represented a very small number of cases, and the inclusion of these cases was felt to be unlikely to impact the overall study outcome. With regard to the examination of the VPS dataset, the main limitation was the voluntary nature of data submission regarding the timing of intubation and/or extubation. As such, in spite of the multiple centers contributing data to the VPS dataset, our evaluation of extubation timing was limited to only those patients with documented intubation and extubation times; thus reducing the effective sample size. Therefore, it is possible that given our reduced sample size, we were underpowered to detect the trend toward reduced duration of mechanical ventilation in patients cared for under an in-house attending model in comparison to those patients who were cared for in units that used other staffing models. Additionally, our study specifically evaluated patients who were mechanically ventilated at the time of postoperative CICU admission. 262
www.pccmjournal.org
Thus, the least complicated patients were less likely to be included in the analysis.
CONCLUSIONS In this study, we found that there are relatively high rates of successful nighttime extubation in the absence of an in-house attending physician. Furthermore, in evaluating national data, it is noteworthy that institutions that have nighttime in-house attending presence do not extubate more patients at night than those that use more traditional staffing models, nor did they demonstrate a consistent reduction in the duration of mechanical ventilation. Although there may be benefits to the presence of an in-house attending physician in the care of patients following congenital heart surgery, the addition of 24/7 in-house attending coverage in a CICU that already uses in-house residents, fellows, or midlevel providers may not significantly reduce the duration of mechanical ventilation.
ACKNOWLEDGMENTS Virtual PICU System (VPS) data were provided from the VPS. No endorsement or editorial restriction of the interpretation of these data or opinions of the authors has been implied or stated. The authors acknowledge Pam Carlock, RN, our surgical database manager, for assistance in accessing the surgical database at Children’s Healthcare of Atlanta.
REFERENCES
1. Cavallazzi R, Marik PE, Hirani A, et al: Association between time of admission to the ICU and mortality: A systematic review and metaanalysis. Chest 2010; 138:68–75 2. Angus DC, Dorman T, Robinson KA, et al: Physician staffing patterns and clinical outcomes in critically ill patients: A systematic review. JAMA 2002; 288:2151–2162 3. Lee SK, Lee DS, Andrews WL, et al: Higher mortality rates among inborn infants admitted to neonatal intensive care units at night. J Pediatr 2003; 143:592–597 4. Wallace DJ, Angus DC, Barnato AE, et al: Nighttime intensivist staffing and mortality among critically ill patients. N Engl J Med 2012; 366:2093–2101 5. Nishisaki A, Pines JM, Lin R, et al: The impact of 24-hr, in-hospital pediatric critical care attending physician presence on process of care and patient outcomes. Crit Care Med 2012; 40:2190–2195 6. Roeleveld PP, Guijt D, Kuijper EJ, et al: Ventilator-associated pneumonia in children after cardiac surgery in The Netherlands. Intensive Care Med 2011; 37:1656–1663 7. Shi S, Zhao Z, Liu X, et al: Perioperative risk factors for prolonged mechanical ventilation following cardiac surgery in neonates and young infants. Chest 2008; 134:768–774 8. Polito A, Patorno E, Costello JM, et al: Perioperative factors associated with prolonged mechanical ventilation after complex congenital heart surgery. Pediatr Crit Care Med 2011; 12:e122–e126 9. Costello JM, Graham DA, Morrow DF, et al: Risk factors for central line-associated bloodstream infection in a pediatric cardiac intensive care unit. Pediatr Crit Care Med 2009; 10:453–459 10. Newburger JW, Wypij D, Bellinger DC, et al: Length of stay after infant heart surgery is related to cognitive outcome at age 8 years. J Pediatr 2003; 143:67–73 11. Jenkins KJ, Gauvreau K, Newburger JW, et al: Consensus-based method for risk adjustment for surgery for congenital heart disease. J Thorac Cardiovasc Surg 2002; 123:110–118 March 2014 • Volume 15 • Number 3
Quality and Safety 12. Bland JM, Altman DG: Transformations, means, and confidence intervals. BMJ 1996; 312:1079 13. Uusaro A, Kari A, Ruokonen E: The effects of ICU admission and discharge times on mortality in Finland. Intensive Care Med 2003; 29:2144–2148 14. Ensminger SA, Morales IJ, Peters SG, et al: The hospital mortality of patients admitted to the ICU on weekends. Chest 2004; 126:1292–1298 15. Luyt CE, Combes A, Aegerter P, et al: Mortality among patients admitted to intensive care units during weekday day shifts compared with “off” hours. Crit Care Med 2007; 35:3–11 16. Gasperino J: The Leapfrog initiative for intensive care unit physician staffing and its impact on intensive care unit performance: A narrative review. Health Policy 2011; 102:223–228 17. Manthous CA: Leapfrog and critical care: Evidence- and reality-based intensive care for the 21st century. Am J Med 2004; 116:188–193 18. Haupt MT, Bekes CE, Brilli RJ, et al; Task Force of the American College of Critical Care Medicine, Society of Critical Care Medicine: Guidelines on critical care services and personnel: Recommendations based on a system of categorization of three levels of care. Crit Care Med 2003; 31:2677–2683 19. Gajic O, Afessa B, Hanson AC, et al: Effect of 24-hour mandatory versus on-demand critical care specialist presence on quality of care and family and provider satisfaction in the intensive care unit of a teaching hospital. Crit Care Med 2008; 36:36–44 20. Hawari FI, Al Najjar TI, Zaru L, et al: The effect of implementing high-intensity intensive care unit staffing model on outcome of critically ill oncology patients. Crit Care Med 2009; 37:1967–1971 21. Parikh A, Huang SA, Murthy P, et al: Quality improvement and cost savings after implementation of the Leapfrog intensive care unit physician staffing standard at a community teaching hospital. Crit Care Med 2012; 40:2754–2759 22. Kumar K, Zarychanski R, Bell DD, et al; Cardiovascular Health Research in Manitoba Investigator Group: Impact of 24-hour in-house
Pediatric Critical Care Medicine
intensivists on a dedicated cardiac surgery intensive care unit. Ann Thorac Surg 2009; 88:1153–1161 23. Lee SK, Lee DS, Andrews WL, et al; Canadian Neonatal Network: Higher mortality rates among inborn infants admitted to neonatal intensive care units at night. J Pediatr 2003; 143:592–597 24. Farnan JM, Johnson JK, Meltzer DO, et al: Resident uncertainty in clinical decision making and impact on patient care: A qualitative study. Qual Saf Health Care 2008; 17:122–126 25. Wernovsky G, Kuijpers M, Van Rossem MC, et al: Postoperative course in the cardiac intensive care unit following the first stage of Norwood reconstruction. Cardiol Young 2007; 17:652–665 26. Cooper DS, Costello JM, Bronicki RA, et al: Current challenges in cardiac intensive care: Optimal strategies for mechanical ventilation and timing of extubation. Cardiol Young 2008; 18(Suppl 3):72–83 27. Karthekeyan RB, Sundar AS, Sulaiman S, et al: Early extubation in tetralogy of Fallot patients after complete repair. Cardiol Young 2011; 21:378–382 28. Winch PD, Nicholson L, Isaacs J, et al: Predictors of successful early extubation following congenital cardiac surgery in neonates and infants. Heart Lung Circ 2009; 18:271–276 29. Yamasaki Y, Shime N, Miyazaki T, et al: Fast-track postoperative care for neonatal cardiac surgery: A single-institute experience. J Anesth 2011; 25:321–329 30. Preisman S, Lembersky H, Yusim Y, et al: A randomized trial of outcomes of anesthetic management directed to very early extubation after cardiac surgery in children. J Cardiothorac Vasc Anesth 2009; 23:348–357 31. Davis S, Worley S, Mee RB, et al: Factors associated with early extubation after cardiac surgery in young children. Pediatr Crit Care Med 2004; 5:63–68 32. Vricella LA, Dearani JA, Gundry SR, et al: Ultra fast track in elective congenital cardiac surgery. Ann Thorac Surg 2000; 69:865–871 33. Mittnacht AJ, Thanjan M, Srivastava S, et al: Extubation in the operating room after congenital heart surgery in children. J Thorac Cardiovasc Surg 2008; 136:88–93
www.pccmjournal.org
263
