- Email: [email protected]
Overview of adverse events related to invasive procedures in the intensive care unit
American Journal of Infection Control 40 (2012) 241-6
Contents lists available at ScienceDirect
American Journal of Infection Control
American Journal of Infection Control
journal homepage: www.ajicjournal.org
Major article
Overview of adverse events related to invasive procedures in the intensive care unit Véronique Pottier MD a, *, Cédric Daubin MD b, Nicolas Lerolle MD, PhD c, Cathy Gaillard MSc d, Gérald Viquesnel MD a, Benoit Plaud MD, PhD a, Jean-Luc Hanouz MD, PhD a, e, Pierre Charbonneau MD, PhD b a
Department of Anesthesia and Surgical Intensive Care, Caen University Hospital, Caen, France Department of Medical Intensive Care, Caen University Hospital, Caen, France Department of Medical Intensive Care, Angers University Hospital, Angers, France d Department of Clinical Research, Caen University Hospital, Caen, France e EA3212, Caen University Hospital, Caen, France b c
Key Words: Iatrogenic disease Medical intensive care Nosocomial infection Ventilator-associated pneumonia Mechanical complications
Background: This study was conducted to determine the frequency, predictors, and clinical impact of adverse events (AEs) related to invasive procedures in the intensive care unit (ICU). Methods: This was a prospective observational study of ICUs in a university hospital. Results: A total of 893 patients requiring invasive procedures were admitted over a 1-year period. Among these, 310 patients (34.7%) experienced a total of 505 AEs. The mean number of AEs per patient was 1.6 1.1 (range, 1-7). Infectious AEs were significantly more frequent than mechanical AEs (60.4% vs 39.6%; P ¼ .01). Factors independently associated with AE occurrence were isolation of multidrug-resistant bacteria at ICU admission, >5 invasive procedures, and ICU length of stay >8 days. Thirty-three AEs (6.5%) resulted in severe clinical impact, including 24 deaths. Ventilator-associated pneumonia (VAP) accounted for 62.5% of the deaths related to AEs. Conclusions: One-third of critically ill patients experienced AEs related to invasive procedures. Severe AEs were associated with 11% of all ICU deaths. VAP was the most frequent AE related to death. An improved assessment of the riskebenefit balance before each invasive procedure and increased efforts to decrease VAP prevalence are needed to reduce AE-related mortality. Copyright Ó 2012 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.
Adverse events (AEs) related to invasive procedures are common in medicine and may result in death. In its 1999 report To Err is Human, the Institute of Medicine estimated that in the United States, AEs (including medical errors of omission and commission) contribute to 44,000-98,000 deaths each year.1 AEs also are responsible for irreversible functional impairment, prolonged and multiple hospitalizations requiring additional invasive therapies leading to risk of further AEs and, consequently, increased health care costs. In France, a 2004 governmental health care agencye sponsored prospective observational study reported an iatrogenic event incidence of 6.6 per 1,000 hospital days.2 In addition, 3%-5% of hospital * Address correspondence to Véronique Pottier, MD, Caen University Hospital, Department of Anaesthesia and Surgical Intensive Care, 14033 Caen Cedex, France. E-mail address: [email protected] (V. Pottier). Conflict of interest: None to report.
stays were due directly to a previous AE, and more than half of these were avoidable. Older age and more severe illness are reportedly the most frequent risk factors for AEs.2-4 Finally, because of the severity of illness and the high number of invasive procedures performed, critically ill patients are at particularly elevated risk for AEs.2 Although previous studies have focused on AEs related to a specific procedure, such as mechanical ventilation5-7 or central venous catheterization,5,8-12 studies assessing the entire range of AEs related to the numerous invasive procedures performed in the ICU setting are scarce. Consequently, the relative importance of various AEs in terms of morbidity and mortality remains poorly assessed. We conducted a prospective observational study to determine the frequency, predictors, and clinical impact of nosocomial infections and mechanical AEs related to common ICU invasive procedures in consecutive critically ill patients admitted to 2 ICUs over a 1-year period.
0196-6553/$36.00 - Copyright Ó 2012 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.ajic.2011.04.005
242
V. Pottier et al. / American Journal of Infection Control 40 (2012) 241-6
Table 1 Predefined invasive procedures and related Aes Invasive procedure
Mechanical AE
Infectious AE
Central venous catheter (including catheter for renal replacement therapy)
Catheter-related venous thrombosis suspected in clinical picture and confirmed by ultrasound Hematoma at the site of insertion Pneumothorax diagnosed on routine chest X-ray Hemothorax diagnosed on routine chest X-ray and confirmed by thoracocentesis
Arterial catheter Chest tube or thoracentesis
Hematoma at the site of insertion Pneumothorax diagnosed on routine chest X-ray Hemothorax diagnosed on routine chest X-ray or CT scan Laryngeal dyspnea, defined as postextubation stridor Tube obstruction, considered only when tube replacement was performed Self-extubation
Endotracheal tube
Tracheotomy tube Urinary tract catheter
Nasogastric tube Cerebral fluid drainage
Tube obstruction, considered only when tube replacement was performed d
Tracheal misplacement with pneumothorax diagnosed on routine chest X-ray Obstruction
METHODS Patients and principle of the study All consecutive adult patients admitted to the 22-bed adult medical ICU and 26-bed adult surgical ICU of Caen University Hospital between January and December 2007 were considered. Patients were included prospectively after undergoing a first invasive procedure in one of these ICUs or in the operating room just before ICU admission. They were followed up for AE occurrence until death or ICU discharge. Invasive ICU procedures were predefined as placement of a urinary tract catheter; an arterial or central venous catheter; a chest, endotracheal, or nasogastric tube; tracheotomy; and a pleural drain or external ventricular drain. All procedures were performed by residents under close supervision by certified physicians, except nasogastric tube and urinary tract catheter placements, which were performed by registered nurses. The AEs related to invasive procedures and their severity were prospectively recorded. AEs related to devices inserted before ICU admission were not considered. All staff members of the 2 ICUs were informed of the study. The study was approved by the local Ethical Committee (CPP Nord Ouest III, CHU de Caen, Caen, France). The committee decided that patient approval was not required, given the observational nature of this prospective study. The database was declared to the French National Commission for computerized files. Data collection and definition of AEs Each patient was screened daily for AEs by the attending intensive care specialist in charge. Twice weekly, the principal
Catheter-related infection Cultures with at least 103 CFU/mL from the catheter tip according to the simplified Brun-Buisson quantitative technique13 Fever or at least 1 positive peripheral blood culture yielding the same species with the same antimicrobial susceptibility as the culture of the catheter tip within 48 hours of catheter removal14 d
VAP15 New or progressive infiltrates on chest X-ray with 2 of the following: Purulent respiratory secretions Fever 38.5 C or hypothermia <36 C Leukocytosis >12.109/L or leukopenia <4.109/L And confirmed by growth in bacterial culture of tracheobronchial aspiration >105 CFU/mL, or >103 CFU/mL on telescopic plugged catheter, or >104 CFU/mL in bronchoalveolar lavage d Urinary tract infection16: leucocytes >106 CFU/mL and bacteria >105 CFU/mL with no more than 2 species of organisms, in the presence of fever >38 C or leucocytes >12.109/L and in the absence of another identifiable source of infection d Meningitis16: 10 nuclear elements/mL, elevated protein, and/or decreased glucose in cerebrospinal fluid and/or organism isolated from cerebrospinal fluid culture
investigator (V.P.) reviewed all patient files to check for misreported AEs. An AE was defined as any complication, mechanical or infectious, related to predefined invasive procedures, regardless of the patient’s underlying condition(s). The definitions and relevant criteria for AEs used in this study are reported in Table 1.13-16 A follow-up file was generated for each AE and its management, including antibiotic treatment for nosocomial infections, at the disposal of the attending intensive care specialist. No specific guidelines were implemented for this study; however, guidelines for ventilator-associated pneumonia (VAP) diagnosis and treatment6,15 and central line insertion, care, and maintenance8,9,11,13 were applied in the 2 ICUs. The severity of each recorded AE was graded into 3 classes at the end of the ICU stay as done in a previous study4: minor, if it had no impact or if it resolved spontaneously; moderate, defined as a prolonged ICU stay or any requirement for additional but routine invasive procedures; and major (fatal or life-threatening), if a specific intensive care therapy was required, such as mechanical ventilation, hemodynamic support, acute surgery, or renal replacement therapy. Grading was first evaluated by the principal investigator (V.P.) and discussed during weekly medical staff meetings with physicians directly involved and not involved in the patient’s care to reach a consensus. In a limited number of cases, no agreement could be reached, and the final decision was made by the principal investigator in consultation with another senior physician not involved in the case. Also recorded were age, sex, underlying diseases, reasons for admission, severity of illness at admission assessed with the new Simplified Acute Physiology Score (SAPS II)17 and the Logistic Organ Dysfunction score (LOD),18 hospital length of stay (LOS) before ICU
V. Pottier et al. / American Journal of Infection Control 40 (2012) 241-6 Table 2 Invasive procedures performed in the 893 critically ill patients studied Type of procedure Endotracheal intubation Central venous catheterization Arterial catheterization Urinary tract catheter Pleural drainage Chest tube Nasogastric tube Tracheotomy External ventricular drain
Number of procedures
Number of patients
931 1,318 835 1,017 31 260 1,149 68 58
746 653 716 863 24 150 767 68 49
admission, isolation of multidrug-resistant bacteria (MRB) found in a screening at ICU admission in our hospital, requirement for mechanical ventilation or renal replacement therapy at any time during the ICU stay, number of invasive procedures during the ICU stay, ICU LOS, and ICU survival. Statistical analysis Quantitative data are expressed as mean SD or median and quartile [first quartile to third quartile], and qualitative data are expressed as percentage. Categorical variables were compared using the c2 test or Fisher’s exact test as appropriate. Quantitative variables were compared using the Student t test for normally distributed data and the Mann-Whitney nonparametric U test when data were not normally distributed (ie, length variables, number of invasive procedures, and LOD score). The association of variables with AE occurrence was assessed by univariate analysis followed by stepwise multivariate logistic regression analysis. The variables tested were age, sex, presence of MRB at ICU admission, severity of illness at admission using the SAPS II17 and the LOD,18 number of invasive procedures, LOS before ICU admission, and ICU LOS. For the latter 4 variables, which were not normally distributed, to have a robust multivariate model, we categorized variables taking the median as a cutoff value. Variables significantly associated (P < .20) with AE occurrence in the univariate analysis were included in the multivariate logistic regression model for association with AE occurrence. Concordance indexes and the Hoshmer-Lemeshow test were used to assess the validity of the multivariate model and calibration, respectively. P values are given together with the odds ratio (OR) and 95% confidence interval (CI). The level of significance was set at P < .05. All statistical analyses were performed with SPSS 14.0 (SPSS Inc, Chicago, IL). RESULTS Patient characteristics and invasive procedures During the study period, 893 of the 1,262 patients (70.8%) admitted to the 2 ICUs (480 in the medical ICU and 413 in the surgical ICU) required at least one invasive procedure during their ICU stay and were included in the analysis. These patients included 575 males (M/F sex ratio,1.80). The mean patient age was 58 17.3 years. At ICU admission, the mean SAPS II score was 45.2 21.2, the median LOD score was 1 [1-2], median hospital LOS before ICU admission was 1 day [1-2], and 16.4% of the patients were MRB-positive. The principal underlying diseases were cardiovascular disorders (509 patients), diabetes mellitus (135 patients), chronic obstructive pulmonary disease (COPD) or asthma (134 patients), cirrhosis (74 patients), cerebrovascular accident (52 patients), and neoplasia (134 patients). Reasons for ICU admission were medical in 54% of cases,
243
unscheduled surgery in 36% of cases, and scheduled surgery in 11% of cases. A total of 5,667 invasive procedures were performed (Table 2). The median number of invasive procedures per patient was 5 [4-7], and the median ICU LOS was 8 days [3-20 days]. There were 670 patients (75%) in the ICU survival group, and 223 patients died. During their ICU stay, 552 (61.8%) critically ill patients required mechanical ventilation, and 201 (22.5%) required renal replacement therapy. Occurrence and clinical impact of AEs related to invasive procedures A total of 310 patients (34.7%) experienced 505 AEs related to invasive procedures; 107 of these patients (34.5%) had 2 or more AEs. The mean number of AEs per patient was 1.6 1.1 (range, 1-7) and the median number of AEs per patient was 1 [1-2]. There was no difference in AE occurrence in terms of frequency or type of AE (ie, infectious and mechanical) between the medical ICU and the surgical ICU (data not shown). Infectious AEs were more frequent than mechanical AEs (305 vs 200; P ¼ .01) (Table 3). VAP was by far the most frequent AE reported (35.3%). In 75% of the cases (n ¼ 381), AEs had either a moderate (n ¼ 348) or a major (n ¼ 33) clinical impact, including 24 deaths, representing 10.7% of all ICU deaths during the study period (Table 3). Overall, the frequency distributions of AEs according to the impact grading (minor, moderate, or major) were not different between mechanical AEs and infectious AEs (P ¼ .87, c2 test). However, VAP was associated with a higher frequency of major AEs (54.5%) than other nosocomial infections (9.1%) or mechanical AEs (36.4%) (P < .001), as shown in Figure 1. Finally, VAP was the main AE responsible for death (15/24). Variables associated with AE and their severity Factors associated with the occurrence of at least one AE related to invasive procedures in the 893 study patients are given in Table 4. In univariate analysis, MRB-positive status at ICU admission, high LOD score, high number of invasive procedures, and LOS before the ICU and in the ICU were significantly associated with the risk of AE occurrence. In multivariate logistic regression analysis, only MRB-positive status, >5 invasive procedures, and ICU LOS >8 days remained significant. The concordance index was 0.818, and the Hoshmer-Lemeshow test was not significant (P ¼ .992), indicating a good ability to differentiate between patients with or without AEs and good calibration, respectively. We examined whether the association of AE with the different parameters tested differed according to the mechanical or infectious subset, and found no difference with the aforementioned general model (data not shown). We failed to identify any parameter associated with the occurrence of severe AEs among our patients, possibly due to the fact that only a few severe AEs could be entered into our statistical models. DISCUSSION In the present prospective study, we assessed the frequency, clinical impact, and factors associated with infectious and mechanical AEs related to commonly performed invasive procedures in the ICU over a 1-year period. We found that 34.7% of patients who underwent invasive procedures experienced at least 1 AE. Major AEs were infrequent (only 6.5% of AEs) but were associated with 24 deaths, representing 11% of all ICU deaths. Infectious AEs occurred more frequently than mechanical AEs (60.4% vs 39.6%). VAP was the main AE reported in terms of both total number of AEs (35% of all AEs) and number of major AEs (55% of
244
V. Pottier et al. / American Journal of Infection Control 40 (2012) 241-6
Table 3 Characteristics and clinical impact of AEs related to invasive procedures
Type of AE
AE severity (n ¼ 505)
Number of AEs (n ¼ 505)
Number of patients (n ¼ 310)
Minor impact
Moderate impact
305 178 86 37 4 200 61 38 33 18 4 9 1 1 26 18 12 0 12 12 10 1 1
199 148 63 32 4 180 48 31 32 16 4 8 1 1 22 15 12 0 12 8 7 1 1
73 0 73 0 0 51 33 0 3 2 0 1 0 0 13 2 0 0 0 0 0 0 0
211 160 12 35 4 137 27 38 29 16 3 8 1 1 13 14 8 0 8 8 6 1 1
Infectious AE VAP Urinary tract infection Catheter-related infection External ventricular drain-related meningitis Mechanical AE Self-extubation Catheter-related venous thrombosis Pneumothorax Postecatheter placement Postintubation Postepleural drainage Postechest tube placement Postenasogastric tube placement Laryngeal dyspnea Hematoma at the site of catheterization Hemothorax Postecatheter placement Postechest tube or pleural drainage Tube obstruction Endotracheal tube Tracheotomy External ventricular drain
Major impact (death) 21 18 1 2
(18) (15) (1) (2) 0 12 (6) 1 0 1 0 1 0 0 0 0 2 4 (2) 0 4 (2) 4 (4) 4 (4*) 0 0
NOTE. Data are presented as number of AE, with number of related-death in brackets for major impact. * Endotracheal tube obstruction resulting in an anoxic encephalopathy.
70
60
58,9 54,5
% AEs
50
46 41,1
39,4
40
VAP
36,4
Other Nosocomial Infection Mechanical Complications
30
20
10
14,6 9,1 0
0 Major (n=33)
Moderate (n=348)
Minor (n=124)
AE Im pact
Fig 1. Distribution of AE severity grouped into AE subsets.
major AEs). In multivariate analysis, only 3 factorsdisolation of MRB at ICU admission, ICU LOS >8 days, and >5 invasive proceduresdwere independently associated with AE occurrence. Despite the vast literature concerning AEs associated with a definite procedure, such as VAP5-7 and catheter-related infection,5,8-12 only a few studies to date have assessed the entire range of AEs related to invasive procedures in ICU patients.19-22 Indeed, defining the scope of AEs to study is controversial, making this objective difficult to define. To clearly delineate the scope of our study, we chose to assess AEs directly related to the most common invasive procedures performed in the ICU. Because of the aforementioned possible variations in the scope of AEs studied, our findings and those of other studies should be compared with
caution. However, our finding of 34.7% of patients with AEs in this setting is in agreement with a recent report using very similar definitions of AEs.21 Conversely, the mean number of AEs per patient was higher in the previous study (2.8, vs 1.6 in our study), possibly due to the larger range of AEs reported.21 In this setting, 3 factors were independently associated with AEs related to invasive procedures: isolation of MRB at ICU admission, ICU LOS >8 days, and >5 invasive procedures. This result is consistent with previous studies.23 Notably, the link between LOS and occurrence of AEs has been established by several studies reporting a direct relationship between length of exposure to the risk (ie, duration of ventilation6 or of vascular catheterization8-11) and the development of infectious complications. Studies focusing
V. Pottier et al. / American Journal of Infection Control 40 (2012) 241-6
245
Table 4 Univariate and multivariate logistic regression analysis of variables associated with AEs
Age, years Male sex MRB-positive status at ICU admission SAPS II LOD score >2 >5 invasive procedures per patient LOS before ICU >2 days ICU LOS >8 days
Univariate analysis
Patients with AEs (n ¼ 310)
Patients free of AEs (n ¼ 583)
P value
59.4 16.4 212 (68.4)* 93 (30.5)* 45.9 18.9 74 (23.9) 223 (71.9) 88 (28.4) 257 (82.9)
57.9 17.7 363 (62.3)y 52 (9.0)y 44.9 22.3 126 (21.6) 168 (28.9) 113 (19.4) 170 (29.4)
.216 .069 <.0001 .504 .441 <.0001 .002 <.0001
OR (95% CI) 1.005 1.311 4.454 1.002 1.137 6.317 1.649 11.638
(0.997-1.013) (0.980-1.757) (3.061-6.481) (0.996-1.009) (0.820-1.578) (4.652-8.576) (1.196-2.273) (8.237-16.442)
Multivariate analysis P value
Adjusted OR (95% CI)
e .518 .001 e e <.0001 .753 <.001
e e 2.159 (1.396-3.338) e e 2.588 (1.804-3.713) e 6.852 (4.688-10.014)
NOTE. Data are presented as mean SD or number (%), as appropriate. * The percentage of patients who experienced AEs was calculated using 310 patients as the denominator. y The percentage of patients who did not experience AEs was calculated using 591 patients as the denominator.
on the relationship between MRB status and AEs are scarce; however, the association between MRB status and infectious AEs is consistent with previous reports that reducing antibiotic exposure reduced the incidence of MRB status and clinical infections.24,25 In contrast, the relationship between MRB status and mechanical AEs is not straightforward. To the best of our knowledge, this finding has not been reported until now. We can only speculate that positive MRB status recorded at ICU admission possibly selects frail patients who are more susceptible to both infectious and mechanical AEs. This point is put into perspective by the lack of association between AE occurrence and initial severity, as assessed by the SAPS II score, indicating that predisposition to AEs may depend more on preexisting factors than on acute illness. This hypothesis merits further study. The predominance of infectious AEs over mechanical AEs (ie, the fact that VAP was the leading AE reported) is in accordance with previous studies.21,22 Unfortunately, based on our findings, we cannot offer a clear explanation of this result; however, we speculate that the probability of AE occurrence increases with increasing duration of exposure to the risk, which is restricted to the time of device insertion for mechanical AEs. Of 310 patients analyzed, 33 experienced major AEs with a significant clinical impact, including 24 deaths (accounting for 11% of all ICU deaths). This result is in contrast with previous studies that used similar AE definitions and reported mortality associated with major AEs ranging from 0.6% to 4.4%.21,22 Differences in selected patients (eg, comorbidities, severity of illness at admission, number of patients with shock or requiring mechanical ventilation and dialysis) could explain this difference, at least in part. However, this finding underlines the impact of iatrogenic complications on mortality in the ICU. Our study cannot answer the question of whether these deaths could have been avoided; however, it may indicate the direction in which efforts should be directed in attempts to improve patient safety. Indeed, we have shown that VAP is the most frequent AE related to death, a result consistent with studies reporting VAP as the primary cause of death among nosocomial infections in the ICU.5-7,21 Therefore, efforts to reduce VAP incidence remain a priority. Toward this end, the implementation of ventilator-weaning protocols and optimal use of sedation in mechanically ventilated patients might be particularly relevant to shorten the duration of mechanical ventilation and VAP.26-29 In our ICUs, sedation was reevaluated twice daily with the objective of sedation, but without a defined protocol. We found that nearly all procedures potentially can be responsible for major AEs and possible death. These findings should serve as a reminder to physicians that every invasive procedure involves a risk of complications. Therefore, the riskebenefit ratio needs to be assessed before proceeding to any invasive procedure. Furthermore, physicians should be more conscientious about receiving
clinical training to improve their skills, thereby decreasing the risk of AEs. The present study has some limitations. First, the patients were followed only until ICU discharge or death, and AEs occurring after ICU discharge were not followed. Although we made a special effort to ensure that patients are discharged to other wards with a minimum of invasive devices, the number of AEs related to invasive procedures likely was underestimated. Second, the specific length of exposure to each invasive procedure was not recorded. But because our main objective was to estimate the overall infectious AE risk, not the risk for each specific procedure, in the interest of simplicity, we deemed that ICU LOS represented a valid estimation of the duration of exposure to risk. Third, we chose to specifically study the spectrum of AEs related to the most common invasive procedures. This implies that some procedures, such as peripheral venous access, or errors relating to medication preparation or administration and adverse effects of correctly administered drugs were not assessed. Furthermore, medical errors, such as missed diagnoses or erroneous treatment decisions, were not taken into account. Further studies are needed to address these important topics. Finally, despite having performed this study in 2 different ICUs with different staffs, we consider the study to be monocentric because these ICUs are located in the same institution and use the same guidelines to manage infectious AEs. Therefore, our results might not be applicable to other institutions with different practices. However, because of the large sample size and a systematic daily screening of AEs, we believe that this study provides useful information regarding the frequency, predictors, and clinical impact of complications of invasive procedures in the ICU. In conclusion, this study shows that the occurrence of AEs related to invasive procedures is a major issue in the ICU with potentially severe clinical consequences, including death. VAP accounted for more than 50% of AE-related deaths, and mechanical complications accounted for 25% of these. But nearly every procedure was found to be occasionally associated with major AEs. Although the ability to avoid severe AEs could not be ascertained, our findings indicate that efforts aimed at reducing the number of invasive procedures and their duration may be advisable to improve patient safety, and that heavy emphasis should be placed on VAP prevention. With this aim in mind, implementation of ventilator-weaning protocols and optimal use of sedation could be particularly relevant. Acknowledgment We thank all of the physicians in the surgical ICU (Philippe Lehoux, Désiré Samba, Claude Jehan, Gilles Leroy, and Pierre Hounfodji) and all physicians in the medical ICU (Damien du
246
V. Pottier et al. / American Journal of Infection Control 40 (2012) 241-6
Cheyron, Michel Ramakers, Nicolas Terzi, and Bruno Bouchet) for their participation in the data collection and the nursing staff, residents, and secretaries of the intensive care units for their cooperation with the protocol. We also thank Valerie FongConstans for polishing the manuscript. References 1. Kohn LT, Corrigan JM, Donaldson MS, editors. To err is human: building a safer health system. Washington [DC]: National Academy Press; 1999. 2. Michel P, Quenon JL, Djihoud A, Tricaud-Vialle S, de Sarasqueta AM, Domecq S. Major adverse events related to the care observed in French hospitals in 2004: results of the study ENEIS. Directorate of Research Studies for Evaluating Data; 2005:398. 3. Darchy B, Le Mière E, Figueredo B, Bavoux E, Cadoux G, Domart Y. Iatrogenic diseases as a reason for admission to the intensive care unit: Incidence, causes, and consequences. Arch Intern Med 1999;159:71-8. 4. Trunet P, Le Gall JR, Lhoste F, Regnier B, Saillard Y, Carlet J, et al. The role of iatrogenic disease in admissions to intensive care. JAMA 1980;244:2617-20. 5. Rosenthal VD, Maki DG, Jamulitrat S, Medeiros EA, Kumar-Todi S, Yepes Gomez D, et al. INICC Members: International Nosocomial Infection Control Consortium (INICC) report, data summary for 2003-2008, issued June 2009. Am J Infect Control 2010;38:95-106. 6. Chastre J, Fagon JY. Ventilator-associated pneumonia. Am J Resp Crit Care Med 2002;165:867-903. 7. Daubin C, Vincent S, Vabret A, du Cheyron D, Parienti JJ, Ramakers M, et al. Nosocomial viral ventilator-associated pneumonia in the intensive care unit. Intensive Care Med 2005;31:1116-22. 8. McGee DC, Gould MK. Preventing complications of central venous catheterization. N Engl J Med 2003;348:1123-33. 9. Parienti JJ, Du Cheyron D, Ramakers M, Malbruny B, Leclercq R, Le Coutour X, et al., members of the NACRE Study Group. Alcoholic povidone-iodine to prevent central venous catheter colonization. Crit Care Med 2004;32:708-13. 10. Parienti JJ, Mégarbane B, Fisher MO, Lautrette A, Gazui N, Marin N, et al., Cathedia Study Group. Catheter dysfunction and dialysis performance according to vascular access among 736 critically ill adults requiring renal replacement therapy: A randomized controlled study. Crit Care Med 2010;38:1-8. 11. Polderman KH, Girbes ARJ. Central venous catheter use: mechanical and infectious complications. Intensive Care Med 2002;28:18-28. 12. Lefrant JY, Muller L, De la Coussaye JE, Prudhomme M, Ripart J, Gouzes C, et al. Risk factors of failure and immediate complication of subclavian vein catheterization in critically ill patients. Intensive Care Med 2002;28:1036-41. 13. Brun-Buisson C, Abrouk F, Legrand P, Huet Y, Larabi S, Rapin M. Diagnosis of central venous cathetererelated sepsis: Critical level of quantitative tip cultures. Arch Intern Med 1987;147:873-7. 14. Parienti JJ, Thirion M, Megarbane B, Souweine B, Ouchikhe A, Polito A, et al., members of the Cathedia Study Group. Femoral vs jugular venous catheterization
15.
16. 17.
18.
19.
20.
21.
22.
23.
24.
25.
26. 27.
28.
29.
and risk of nosocomial events in adults requiring acute renal replacement therapy. JAMA 2008;299:2413-22. Rello J, Paiva JA, Baraibar J, Barcenilla F, Bodi M, Castander D, et al. International conference for the development of consensus on the diagnosis and treatment of ventilator-associated pneumonia. Chest 2001;120:955-70. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections. Am J Infect Control 1988;16:128-40. Le Gall JR, Lemeshow S, Saulnier F. A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA 1993; 270:2957-63. Le Gall JR, Klar J, Lemeshow S, Saulnier F, Alberti C, Artigas A, Teres D., ICU Scoring Group. The Logistic Organ Dysfunction system: a new way to assess organ dysfunction in the intensive care unit. JAMA 1996;276:802-10. Rothschild JM, Landrigan CP, Cronin JW, Kaushal R, Lockley SW, Burdick E, et al. The Critical Care Safety Study: The incidence and nature of adverse events and serious medical errors in intensive care. Crit Care Med 2005;33:1694-700. Lehmann LS, Puopolo AL, Shaykevich S, Brennan TA. Iatrogenic events resulting in intensive care admission: Frequency, cause, and disclosure to patients and institutions. Am J Med 2005;118:409-13. Garrouste Orgeas M, Timsit JF, Soufir L, Tafflet M, Adrie C, Philippart F, et al., on behalf of the Outcomerea Study Group. Impact of adverse events on outcomes in ICU patients. Crit Care Med 2008;36:1-7. Giraud T, Dhainaut JF, Vaxelaire JF, Joseph T, Journois D, Bleichner G, et al. Iatrogenic complications in adult intensive care units: A prospective twocenter study. Crit Care Med 1993;21:40-51. Valentin A, Capuzzo M, Guidet B, Moreno RP, Dolanski L, Bauer P, et al., Research Group on Quality Improvement of European Society of Intensive Care Medicine. Patient safety in intensive care: results from the multinational Sentinel Events Evaluation (SEE) study. Intensive Care Med 2006;32:1591-8. Charbonneau P, Parienti JJ, Thibon P, Ramakers M, Daubin C, Du Cheyron D, et al. Fluoroquinolone use and méthicillin-resistant Staphylococcus aureus isolation rates in hospitalized patients. Clin Infect Dis 2006;42:778-84. Tacconelli E, De Angelis G, Cataldo MA, Pozzi E, Cauda R. Does antibiotic exposure increase the risk of methicillin-resistant Staphylococcus aureus (MRSA) isolation? A systematic review and meta-analysis. J Antimicrob Chemother 2008;61:26-38. Kollef MH. Prevention of hospital-associated pneumonia and ventilatorassociated pneumonia. Crit Care Med 2004;32:1396-405. Brook AD, Ahrens TS, Schaiff R, Prentice D, Sherman G, Shannon W, et al. Effect of nursing-implemented sedation protocol on the duration of mechanical ventilation. Crit Care Med 1999;27:2609-15. Schweickert WD, Gehlbach BK, Pohlman AS, Hall JB, Kress JP. Daily interruption of sedative infusions and complications of critical illness in mechanically ventilated patients. Crit Care Med 2004;32:1272-6. Girard TD, Kress JP, Fuchs BD, Thomason JWW, Schweickert WD, Pun BT, et al. Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically ventilated patients in intensive care (awakening and breathing controlled trial): A randomised controlled trial. Lancet 2008;371: 126-34.
Contents lists available at ScienceDirect
American Journal of Infection Control
American Journal of Infection Control
journal homepage: www.ajicjournal.org
Major article
Overview of adverse events related to invasive procedures in the intensive care unit Véronique Pottier MD a, *, Cédric Daubin MD b, Nicolas Lerolle MD, PhD c, Cathy Gaillard MSc d, Gérald Viquesnel MD a, Benoit Plaud MD, PhD a, Jean-Luc Hanouz MD, PhD a, e, Pierre Charbonneau MD, PhD b a
Department of Anesthesia and Surgical Intensive Care, Caen University Hospital, Caen, France Department of Medical Intensive Care, Caen University Hospital, Caen, France Department of Medical Intensive Care, Angers University Hospital, Angers, France d Department of Clinical Research, Caen University Hospital, Caen, France e EA3212, Caen University Hospital, Caen, France b c
Key Words: Iatrogenic disease Medical intensive care Nosocomial infection Ventilator-associated pneumonia Mechanical complications
Background: This study was conducted to determine the frequency, predictors, and clinical impact of adverse events (AEs) related to invasive procedures in the intensive care unit (ICU). Methods: This was a prospective observational study of ICUs in a university hospital. Results: A total of 893 patients requiring invasive procedures were admitted over a 1-year period. Among these, 310 patients (34.7%) experienced a total of 505 AEs. The mean number of AEs per patient was 1.6 1.1 (range, 1-7). Infectious AEs were significantly more frequent than mechanical AEs (60.4% vs 39.6%; P ¼ .01). Factors independently associated with AE occurrence were isolation of multidrug-resistant bacteria at ICU admission, >5 invasive procedures, and ICU length of stay >8 days. Thirty-three AEs (6.5%) resulted in severe clinical impact, including 24 deaths. Ventilator-associated pneumonia (VAP) accounted for 62.5% of the deaths related to AEs. Conclusions: One-third of critically ill patients experienced AEs related to invasive procedures. Severe AEs were associated with 11% of all ICU deaths. VAP was the most frequent AE related to death. An improved assessment of the riskebenefit balance before each invasive procedure and increased efforts to decrease VAP prevalence are needed to reduce AE-related mortality. Copyright Ó 2012 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.
Adverse events (AEs) related to invasive procedures are common in medicine and may result in death. In its 1999 report To Err is Human, the Institute of Medicine estimated that in the United States, AEs (including medical errors of omission and commission) contribute to 44,000-98,000 deaths each year.1 AEs also are responsible for irreversible functional impairment, prolonged and multiple hospitalizations requiring additional invasive therapies leading to risk of further AEs and, consequently, increased health care costs. In France, a 2004 governmental health care agencye sponsored prospective observational study reported an iatrogenic event incidence of 6.6 per 1,000 hospital days.2 In addition, 3%-5% of hospital * Address correspondence to Véronique Pottier, MD, Caen University Hospital, Department of Anaesthesia and Surgical Intensive Care, 14033 Caen Cedex, France. E-mail address: [email protected] (V. Pottier). Conflict of interest: None to report.
stays were due directly to a previous AE, and more than half of these were avoidable. Older age and more severe illness are reportedly the most frequent risk factors for AEs.2-4 Finally, because of the severity of illness and the high number of invasive procedures performed, critically ill patients are at particularly elevated risk for AEs.2 Although previous studies have focused on AEs related to a specific procedure, such as mechanical ventilation5-7 or central venous catheterization,5,8-12 studies assessing the entire range of AEs related to the numerous invasive procedures performed in the ICU setting are scarce. Consequently, the relative importance of various AEs in terms of morbidity and mortality remains poorly assessed. We conducted a prospective observational study to determine the frequency, predictors, and clinical impact of nosocomial infections and mechanical AEs related to common ICU invasive procedures in consecutive critically ill patients admitted to 2 ICUs over a 1-year period.
0196-6553/$36.00 - Copyright Ó 2012 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.ajic.2011.04.005
242
V. Pottier et al. / American Journal of Infection Control 40 (2012) 241-6
Table 1 Predefined invasive procedures and related Aes Invasive procedure
Mechanical AE
Infectious AE
Central venous catheter (including catheter for renal replacement therapy)
Catheter-related venous thrombosis suspected in clinical picture and confirmed by ultrasound Hematoma at the site of insertion Pneumothorax diagnosed on routine chest X-ray Hemothorax diagnosed on routine chest X-ray and confirmed by thoracocentesis
Arterial catheter Chest tube or thoracentesis
Hematoma at the site of insertion Pneumothorax diagnosed on routine chest X-ray Hemothorax diagnosed on routine chest X-ray or CT scan Laryngeal dyspnea, defined as postextubation stridor Tube obstruction, considered only when tube replacement was performed Self-extubation
Endotracheal tube
Tracheotomy tube Urinary tract catheter
Nasogastric tube Cerebral fluid drainage
Tube obstruction, considered only when tube replacement was performed d
Tracheal misplacement with pneumothorax diagnosed on routine chest X-ray Obstruction
METHODS Patients and principle of the study All consecutive adult patients admitted to the 22-bed adult medical ICU and 26-bed adult surgical ICU of Caen University Hospital between January and December 2007 were considered. Patients were included prospectively after undergoing a first invasive procedure in one of these ICUs or in the operating room just before ICU admission. They were followed up for AE occurrence until death or ICU discharge. Invasive ICU procedures were predefined as placement of a urinary tract catheter; an arterial or central venous catheter; a chest, endotracheal, or nasogastric tube; tracheotomy; and a pleural drain or external ventricular drain. All procedures were performed by residents under close supervision by certified physicians, except nasogastric tube and urinary tract catheter placements, which were performed by registered nurses. The AEs related to invasive procedures and their severity were prospectively recorded. AEs related to devices inserted before ICU admission were not considered. All staff members of the 2 ICUs were informed of the study. The study was approved by the local Ethical Committee (CPP Nord Ouest III, CHU de Caen, Caen, France). The committee decided that patient approval was not required, given the observational nature of this prospective study. The database was declared to the French National Commission for computerized files. Data collection and definition of AEs Each patient was screened daily for AEs by the attending intensive care specialist in charge. Twice weekly, the principal
Catheter-related infection Cultures with at least 103 CFU/mL from the catheter tip according to the simplified Brun-Buisson quantitative technique13 Fever or at least 1 positive peripheral blood culture yielding the same species with the same antimicrobial susceptibility as the culture of the catheter tip within 48 hours of catheter removal14 d
VAP15 New or progressive infiltrates on chest X-ray with 2 of the following: Purulent respiratory secretions Fever 38.5 C or hypothermia <36 C Leukocytosis >12.109/L or leukopenia <4.109/L And confirmed by growth in bacterial culture of tracheobronchial aspiration >105 CFU/mL, or >103 CFU/mL on telescopic plugged catheter, or >104 CFU/mL in bronchoalveolar lavage d Urinary tract infection16: leucocytes >106 CFU/mL and bacteria >105 CFU/mL with no more than 2 species of organisms, in the presence of fever >38 C or leucocytes >12.109/L and in the absence of another identifiable source of infection d Meningitis16: 10 nuclear elements/mL, elevated protein, and/or decreased glucose in cerebrospinal fluid and/or organism isolated from cerebrospinal fluid culture
investigator (V.P.) reviewed all patient files to check for misreported AEs. An AE was defined as any complication, mechanical or infectious, related to predefined invasive procedures, regardless of the patient’s underlying condition(s). The definitions and relevant criteria for AEs used in this study are reported in Table 1.13-16 A follow-up file was generated for each AE and its management, including antibiotic treatment for nosocomial infections, at the disposal of the attending intensive care specialist. No specific guidelines were implemented for this study; however, guidelines for ventilator-associated pneumonia (VAP) diagnosis and treatment6,15 and central line insertion, care, and maintenance8,9,11,13 were applied in the 2 ICUs. The severity of each recorded AE was graded into 3 classes at the end of the ICU stay as done in a previous study4: minor, if it had no impact or if it resolved spontaneously; moderate, defined as a prolonged ICU stay or any requirement for additional but routine invasive procedures; and major (fatal or life-threatening), if a specific intensive care therapy was required, such as mechanical ventilation, hemodynamic support, acute surgery, or renal replacement therapy. Grading was first evaluated by the principal investigator (V.P.) and discussed during weekly medical staff meetings with physicians directly involved and not involved in the patient’s care to reach a consensus. In a limited number of cases, no agreement could be reached, and the final decision was made by the principal investigator in consultation with another senior physician not involved in the case. Also recorded were age, sex, underlying diseases, reasons for admission, severity of illness at admission assessed with the new Simplified Acute Physiology Score (SAPS II)17 and the Logistic Organ Dysfunction score (LOD),18 hospital length of stay (LOS) before ICU
V. Pottier et al. / American Journal of Infection Control 40 (2012) 241-6 Table 2 Invasive procedures performed in the 893 critically ill patients studied Type of procedure Endotracheal intubation Central venous catheterization Arterial catheterization Urinary tract catheter Pleural drainage Chest tube Nasogastric tube Tracheotomy External ventricular drain
Number of procedures
Number of patients
931 1,318 835 1,017 31 260 1,149 68 58
746 653 716 863 24 150 767 68 49
admission, isolation of multidrug-resistant bacteria (MRB) found in a screening at ICU admission in our hospital, requirement for mechanical ventilation or renal replacement therapy at any time during the ICU stay, number of invasive procedures during the ICU stay, ICU LOS, and ICU survival. Statistical analysis Quantitative data are expressed as mean SD or median and quartile [first quartile to third quartile], and qualitative data are expressed as percentage. Categorical variables were compared using the c2 test or Fisher’s exact test as appropriate. Quantitative variables were compared using the Student t test for normally distributed data and the Mann-Whitney nonparametric U test when data were not normally distributed (ie, length variables, number of invasive procedures, and LOD score). The association of variables with AE occurrence was assessed by univariate analysis followed by stepwise multivariate logistic regression analysis. The variables tested were age, sex, presence of MRB at ICU admission, severity of illness at admission using the SAPS II17 and the LOD,18 number of invasive procedures, LOS before ICU admission, and ICU LOS. For the latter 4 variables, which were not normally distributed, to have a robust multivariate model, we categorized variables taking the median as a cutoff value. Variables significantly associated (P < .20) with AE occurrence in the univariate analysis were included in the multivariate logistic regression model for association with AE occurrence. Concordance indexes and the Hoshmer-Lemeshow test were used to assess the validity of the multivariate model and calibration, respectively. P values are given together with the odds ratio (OR) and 95% confidence interval (CI). The level of significance was set at P < .05. All statistical analyses were performed with SPSS 14.0 (SPSS Inc, Chicago, IL). RESULTS Patient characteristics and invasive procedures During the study period, 893 of the 1,262 patients (70.8%) admitted to the 2 ICUs (480 in the medical ICU and 413 in the surgical ICU) required at least one invasive procedure during their ICU stay and were included in the analysis. These patients included 575 males (M/F sex ratio,1.80). The mean patient age was 58 17.3 years. At ICU admission, the mean SAPS II score was 45.2 21.2, the median LOD score was 1 [1-2], median hospital LOS before ICU admission was 1 day [1-2], and 16.4% of the patients were MRB-positive. The principal underlying diseases were cardiovascular disorders (509 patients), diabetes mellitus (135 patients), chronic obstructive pulmonary disease (COPD) or asthma (134 patients), cirrhosis (74 patients), cerebrovascular accident (52 patients), and neoplasia (134 patients). Reasons for ICU admission were medical in 54% of cases,
243
unscheduled surgery in 36% of cases, and scheduled surgery in 11% of cases. A total of 5,667 invasive procedures were performed (Table 2). The median number of invasive procedures per patient was 5 [4-7], and the median ICU LOS was 8 days [3-20 days]. There were 670 patients (75%) in the ICU survival group, and 223 patients died. During their ICU stay, 552 (61.8%) critically ill patients required mechanical ventilation, and 201 (22.5%) required renal replacement therapy. Occurrence and clinical impact of AEs related to invasive procedures A total of 310 patients (34.7%) experienced 505 AEs related to invasive procedures; 107 of these patients (34.5%) had 2 or more AEs. The mean number of AEs per patient was 1.6 1.1 (range, 1-7) and the median number of AEs per patient was 1 [1-2]. There was no difference in AE occurrence in terms of frequency or type of AE (ie, infectious and mechanical) between the medical ICU and the surgical ICU (data not shown). Infectious AEs were more frequent than mechanical AEs (305 vs 200; P ¼ .01) (Table 3). VAP was by far the most frequent AE reported (35.3%). In 75% of the cases (n ¼ 381), AEs had either a moderate (n ¼ 348) or a major (n ¼ 33) clinical impact, including 24 deaths, representing 10.7% of all ICU deaths during the study period (Table 3). Overall, the frequency distributions of AEs according to the impact grading (minor, moderate, or major) were not different between mechanical AEs and infectious AEs (P ¼ .87, c2 test). However, VAP was associated with a higher frequency of major AEs (54.5%) than other nosocomial infections (9.1%) or mechanical AEs (36.4%) (P < .001), as shown in Figure 1. Finally, VAP was the main AE responsible for death (15/24). Variables associated with AE and their severity Factors associated with the occurrence of at least one AE related to invasive procedures in the 893 study patients are given in Table 4. In univariate analysis, MRB-positive status at ICU admission, high LOD score, high number of invasive procedures, and LOS before the ICU and in the ICU were significantly associated with the risk of AE occurrence. In multivariate logistic regression analysis, only MRB-positive status, >5 invasive procedures, and ICU LOS >8 days remained significant. The concordance index was 0.818, and the Hoshmer-Lemeshow test was not significant (P ¼ .992), indicating a good ability to differentiate between patients with or without AEs and good calibration, respectively. We examined whether the association of AE with the different parameters tested differed according to the mechanical or infectious subset, and found no difference with the aforementioned general model (data not shown). We failed to identify any parameter associated with the occurrence of severe AEs among our patients, possibly due to the fact that only a few severe AEs could be entered into our statistical models. DISCUSSION In the present prospective study, we assessed the frequency, clinical impact, and factors associated with infectious and mechanical AEs related to commonly performed invasive procedures in the ICU over a 1-year period. We found that 34.7% of patients who underwent invasive procedures experienced at least 1 AE. Major AEs were infrequent (only 6.5% of AEs) but were associated with 24 deaths, representing 11% of all ICU deaths. Infectious AEs occurred more frequently than mechanical AEs (60.4% vs 39.6%). VAP was the main AE reported in terms of both total number of AEs (35% of all AEs) and number of major AEs (55% of
244
V. Pottier et al. / American Journal of Infection Control 40 (2012) 241-6
Table 3 Characteristics and clinical impact of AEs related to invasive procedures
Type of AE
AE severity (n ¼ 505)
Number of AEs (n ¼ 505)
Number of patients (n ¼ 310)
Minor impact
Moderate impact
305 178 86 37 4 200 61 38 33 18 4 9 1 1 26 18 12 0 12 12 10 1 1
199 148 63 32 4 180 48 31 32 16 4 8 1 1 22 15 12 0 12 8 7 1 1
73 0 73 0 0 51 33 0 3 2 0 1 0 0 13 2 0 0 0 0 0 0 0
211 160 12 35 4 137 27 38 29 16 3 8 1 1 13 14 8 0 8 8 6 1 1
Infectious AE VAP Urinary tract infection Catheter-related infection External ventricular drain-related meningitis Mechanical AE Self-extubation Catheter-related venous thrombosis Pneumothorax Postecatheter placement Postintubation Postepleural drainage Postechest tube placement Postenasogastric tube placement Laryngeal dyspnea Hematoma at the site of catheterization Hemothorax Postecatheter placement Postechest tube or pleural drainage Tube obstruction Endotracheal tube Tracheotomy External ventricular drain
Major impact (death) 21 18 1 2
(18) (15) (1) (2) 0 12 (6) 1 0 1 0 1 0 0 0 0 2 4 (2) 0 4 (2) 4 (4) 4 (4*) 0 0
NOTE. Data are presented as number of AE, with number of related-death in brackets for major impact. * Endotracheal tube obstruction resulting in an anoxic encephalopathy.
70
60
58,9 54,5
% AEs
50
46 41,1
39,4
40
VAP
36,4
Other Nosocomial Infection Mechanical Complications
30
20
10
14,6 9,1 0
0 Major (n=33)
Moderate (n=348)
Minor (n=124)
AE Im pact
Fig 1. Distribution of AE severity grouped into AE subsets.
major AEs). In multivariate analysis, only 3 factorsdisolation of MRB at ICU admission, ICU LOS >8 days, and >5 invasive proceduresdwere independently associated with AE occurrence. Despite the vast literature concerning AEs associated with a definite procedure, such as VAP5-7 and catheter-related infection,5,8-12 only a few studies to date have assessed the entire range of AEs related to invasive procedures in ICU patients.19-22 Indeed, defining the scope of AEs to study is controversial, making this objective difficult to define. To clearly delineate the scope of our study, we chose to assess AEs directly related to the most common invasive procedures performed in the ICU. Because of the aforementioned possible variations in the scope of AEs studied, our findings and those of other studies should be compared with
caution. However, our finding of 34.7% of patients with AEs in this setting is in agreement with a recent report using very similar definitions of AEs.21 Conversely, the mean number of AEs per patient was higher in the previous study (2.8, vs 1.6 in our study), possibly due to the larger range of AEs reported.21 In this setting, 3 factors were independently associated with AEs related to invasive procedures: isolation of MRB at ICU admission, ICU LOS >8 days, and >5 invasive procedures. This result is consistent with previous studies.23 Notably, the link between LOS and occurrence of AEs has been established by several studies reporting a direct relationship between length of exposure to the risk (ie, duration of ventilation6 or of vascular catheterization8-11) and the development of infectious complications. Studies focusing
V. Pottier et al. / American Journal of Infection Control 40 (2012) 241-6
245
Table 4 Univariate and multivariate logistic regression analysis of variables associated with AEs
Age, years Male sex MRB-positive status at ICU admission SAPS II LOD score >2 >5 invasive procedures per patient LOS before ICU >2 days ICU LOS >8 days
Univariate analysis
Patients with AEs (n ¼ 310)
Patients free of AEs (n ¼ 583)
P value
59.4 16.4 212 (68.4)* 93 (30.5)* 45.9 18.9 74 (23.9) 223 (71.9) 88 (28.4) 257 (82.9)
57.9 17.7 363 (62.3)y 52 (9.0)y 44.9 22.3 126 (21.6) 168 (28.9) 113 (19.4) 170 (29.4)
.216 .069 <.0001 .504 .441 <.0001 .002 <.0001
OR (95% CI) 1.005 1.311 4.454 1.002 1.137 6.317 1.649 11.638
(0.997-1.013) (0.980-1.757) (3.061-6.481) (0.996-1.009) (0.820-1.578) (4.652-8.576) (1.196-2.273) (8.237-16.442)
Multivariate analysis P value
Adjusted OR (95% CI)
e .518 .001 e e <.0001 .753 <.001
e e 2.159 (1.396-3.338) e e 2.588 (1.804-3.713) e 6.852 (4.688-10.014)
NOTE. Data are presented as mean SD or number (%), as appropriate. * The percentage of patients who experienced AEs was calculated using 310 patients as the denominator. y The percentage of patients who did not experience AEs was calculated using 591 patients as the denominator.
on the relationship between MRB status and AEs are scarce; however, the association between MRB status and infectious AEs is consistent with previous reports that reducing antibiotic exposure reduced the incidence of MRB status and clinical infections.24,25 In contrast, the relationship between MRB status and mechanical AEs is not straightforward. To the best of our knowledge, this finding has not been reported until now. We can only speculate that positive MRB status recorded at ICU admission possibly selects frail patients who are more susceptible to both infectious and mechanical AEs. This point is put into perspective by the lack of association between AE occurrence and initial severity, as assessed by the SAPS II score, indicating that predisposition to AEs may depend more on preexisting factors than on acute illness. This hypothesis merits further study. The predominance of infectious AEs over mechanical AEs (ie, the fact that VAP was the leading AE reported) is in accordance with previous studies.21,22 Unfortunately, based on our findings, we cannot offer a clear explanation of this result; however, we speculate that the probability of AE occurrence increases with increasing duration of exposure to the risk, which is restricted to the time of device insertion for mechanical AEs. Of 310 patients analyzed, 33 experienced major AEs with a significant clinical impact, including 24 deaths (accounting for 11% of all ICU deaths). This result is in contrast with previous studies that used similar AE definitions and reported mortality associated with major AEs ranging from 0.6% to 4.4%.21,22 Differences in selected patients (eg, comorbidities, severity of illness at admission, number of patients with shock or requiring mechanical ventilation and dialysis) could explain this difference, at least in part. However, this finding underlines the impact of iatrogenic complications on mortality in the ICU. Our study cannot answer the question of whether these deaths could have been avoided; however, it may indicate the direction in which efforts should be directed in attempts to improve patient safety. Indeed, we have shown that VAP is the most frequent AE related to death, a result consistent with studies reporting VAP as the primary cause of death among nosocomial infections in the ICU.5-7,21 Therefore, efforts to reduce VAP incidence remain a priority. Toward this end, the implementation of ventilator-weaning protocols and optimal use of sedation in mechanically ventilated patients might be particularly relevant to shorten the duration of mechanical ventilation and VAP.26-29 In our ICUs, sedation was reevaluated twice daily with the objective of sedation, but without a defined protocol. We found that nearly all procedures potentially can be responsible for major AEs and possible death. These findings should serve as a reminder to physicians that every invasive procedure involves a risk of complications. Therefore, the riskebenefit ratio needs to be assessed before proceeding to any invasive procedure. Furthermore, physicians should be more conscientious about receiving
clinical training to improve their skills, thereby decreasing the risk of AEs. The present study has some limitations. First, the patients were followed only until ICU discharge or death, and AEs occurring after ICU discharge were not followed. Although we made a special effort to ensure that patients are discharged to other wards with a minimum of invasive devices, the number of AEs related to invasive procedures likely was underestimated. Second, the specific length of exposure to each invasive procedure was not recorded. But because our main objective was to estimate the overall infectious AE risk, not the risk for each specific procedure, in the interest of simplicity, we deemed that ICU LOS represented a valid estimation of the duration of exposure to risk. Third, we chose to specifically study the spectrum of AEs related to the most common invasive procedures. This implies that some procedures, such as peripheral venous access, or errors relating to medication preparation or administration and adverse effects of correctly administered drugs were not assessed. Furthermore, medical errors, such as missed diagnoses or erroneous treatment decisions, were not taken into account. Further studies are needed to address these important topics. Finally, despite having performed this study in 2 different ICUs with different staffs, we consider the study to be monocentric because these ICUs are located in the same institution and use the same guidelines to manage infectious AEs. Therefore, our results might not be applicable to other institutions with different practices. However, because of the large sample size and a systematic daily screening of AEs, we believe that this study provides useful information regarding the frequency, predictors, and clinical impact of complications of invasive procedures in the ICU. In conclusion, this study shows that the occurrence of AEs related to invasive procedures is a major issue in the ICU with potentially severe clinical consequences, including death. VAP accounted for more than 50% of AE-related deaths, and mechanical complications accounted for 25% of these. But nearly every procedure was found to be occasionally associated with major AEs. Although the ability to avoid severe AEs could not be ascertained, our findings indicate that efforts aimed at reducing the number of invasive procedures and their duration may be advisable to improve patient safety, and that heavy emphasis should be placed on VAP prevention. With this aim in mind, implementation of ventilator-weaning protocols and optimal use of sedation could be particularly relevant. Acknowledgment We thank all of the physicians in the surgical ICU (Philippe Lehoux, Désiré Samba, Claude Jehan, Gilles Leroy, and Pierre Hounfodji) and all physicians in the medical ICU (Damien du
246
V. Pottier et al. / American Journal of Infection Control 40 (2012) 241-6
Cheyron, Michel Ramakers, Nicolas Terzi, and Bruno Bouchet) for their participation in the data collection and the nursing staff, residents, and secretaries of the intensive care units for their cooperation with the protocol. We also thank Valerie FongConstans for polishing the manuscript. References 1. Kohn LT, Corrigan JM, Donaldson MS, editors. To err is human: building a safer health system. Washington [DC]: National Academy Press; 1999. 2. Michel P, Quenon JL, Djihoud A, Tricaud-Vialle S, de Sarasqueta AM, Domecq S. Major adverse events related to the care observed in French hospitals in 2004: results of the study ENEIS. Directorate of Research Studies for Evaluating Data; 2005:398. 3. Darchy B, Le Mière E, Figueredo B, Bavoux E, Cadoux G, Domart Y. Iatrogenic diseases as a reason for admission to the intensive care unit: Incidence, causes, and consequences. Arch Intern Med 1999;159:71-8. 4. Trunet P, Le Gall JR, Lhoste F, Regnier B, Saillard Y, Carlet J, et al. The role of iatrogenic disease in admissions to intensive care. JAMA 1980;244:2617-20. 5. Rosenthal VD, Maki DG, Jamulitrat S, Medeiros EA, Kumar-Todi S, Yepes Gomez D, et al. INICC Members: International Nosocomial Infection Control Consortium (INICC) report, data summary for 2003-2008, issued June 2009. Am J Infect Control 2010;38:95-106. 6. Chastre J, Fagon JY. Ventilator-associated pneumonia. Am J Resp Crit Care Med 2002;165:867-903. 7. Daubin C, Vincent S, Vabret A, du Cheyron D, Parienti JJ, Ramakers M, et al. Nosocomial viral ventilator-associated pneumonia in the intensive care unit. Intensive Care Med 2005;31:1116-22. 8. McGee DC, Gould MK. Preventing complications of central venous catheterization. N Engl J Med 2003;348:1123-33. 9. Parienti JJ, Du Cheyron D, Ramakers M, Malbruny B, Leclercq R, Le Coutour X, et al., members of the NACRE Study Group. Alcoholic povidone-iodine to prevent central venous catheter colonization. Crit Care Med 2004;32:708-13. 10. Parienti JJ, Mégarbane B, Fisher MO, Lautrette A, Gazui N, Marin N, et al., Cathedia Study Group. Catheter dysfunction and dialysis performance according to vascular access among 736 critically ill adults requiring renal replacement therapy: A randomized controlled study. Crit Care Med 2010;38:1-8. 11. Polderman KH, Girbes ARJ. Central venous catheter use: mechanical and infectious complications. Intensive Care Med 2002;28:18-28. 12. Lefrant JY, Muller L, De la Coussaye JE, Prudhomme M, Ripart J, Gouzes C, et al. Risk factors of failure and immediate complication of subclavian vein catheterization in critically ill patients. Intensive Care Med 2002;28:1036-41. 13. Brun-Buisson C, Abrouk F, Legrand P, Huet Y, Larabi S, Rapin M. Diagnosis of central venous cathetererelated sepsis: Critical level of quantitative tip cultures. Arch Intern Med 1987;147:873-7. 14. Parienti JJ, Thirion M, Megarbane B, Souweine B, Ouchikhe A, Polito A, et al., members of the Cathedia Study Group. Femoral vs jugular venous catheterization
15.
16. 17.
18.
19.
20.
21.
22.
23.
24.
25.
26. 27.
28.
29.
and risk of nosocomial events in adults requiring acute renal replacement therapy. JAMA 2008;299:2413-22. Rello J, Paiva JA, Baraibar J, Barcenilla F, Bodi M, Castander D, et al. International conference for the development of consensus on the diagnosis and treatment of ventilator-associated pneumonia. Chest 2001;120:955-70. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections. Am J Infect Control 1988;16:128-40. Le Gall JR, Lemeshow S, Saulnier F. A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA 1993; 270:2957-63. Le Gall JR, Klar J, Lemeshow S, Saulnier F, Alberti C, Artigas A, Teres D., ICU Scoring Group. The Logistic Organ Dysfunction system: a new way to assess organ dysfunction in the intensive care unit. JAMA 1996;276:802-10. Rothschild JM, Landrigan CP, Cronin JW, Kaushal R, Lockley SW, Burdick E, et al. The Critical Care Safety Study: The incidence and nature of adverse events and serious medical errors in intensive care. Crit Care Med 2005;33:1694-700. Lehmann LS, Puopolo AL, Shaykevich S, Brennan TA. Iatrogenic events resulting in intensive care admission: Frequency, cause, and disclosure to patients and institutions. Am J Med 2005;118:409-13. Garrouste Orgeas M, Timsit JF, Soufir L, Tafflet M, Adrie C, Philippart F, et al., on behalf of the Outcomerea Study Group. Impact of adverse events on outcomes in ICU patients. Crit Care Med 2008;36:1-7. Giraud T, Dhainaut JF, Vaxelaire JF, Joseph T, Journois D, Bleichner G, et al. Iatrogenic complications in adult intensive care units: A prospective twocenter study. Crit Care Med 1993;21:40-51. Valentin A, Capuzzo M, Guidet B, Moreno RP, Dolanski L, Bauer P, et al., Research Group on Quality Improvement of European Society of Intensive Care Medicine. Patient safety in intensive care: results from the multinational Sentinel Events Evaluation (SEE) study. Intensive Care Med 2006;32:1591-8. Charbonneau P, Parienti JJ, Thibon P, Ramakers M, Daubin C, Du Cheyron D, et al. Fluoroquinolone use and méthicillin-resistant Staphylococcus aureus isolation rates in hospitalized patients. Clin Infect Dis 2006;42:778-84. Tacconelli E, De Angelis G, Cataldo MA, Pozzi E, Cauda R. Does antibiotic exposure increase the risk of methicillin-resistant Staphylococcus aureus (MRSA) isolation? A systematic review and meta-analysis. J Antimicrob Chemother 2008;61:26-38. Kollef MH. Prevention of hospital-associated pneumonia and ventilatorassociated pneumonia. Crit Care Med 2004;32:1396-405. Brook AD, Ahrens TS, Schaiff R, Prentice D, Sherman G, Shannon W, et al. Effect of nursing-implemented sedation protocol on the duration of mechanical ventilation. Crit Care Med 1999;27:2609-15. Schweickert WD, Gehlbach BK, Pohlman AS, Hall JB, Kress JP. Daily interruption of sedative infusions and complications of critical illness in mechanically ventilated patients. Crit Care Med 2004;32:1272-6. Girard TD, Kress JP, Fuchs BD, Thomason JWW, Schweickert WD, Pun BT, et al. Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically ventilated patients in intensive care (awakening and breathing controlled trial): A randomised controlled trial. Lancet 2008;371: 126-34.