Long-term outcome and quality of life of patients requiring multidisciplinary intensive care unit admission after cardiac operations

LONG-TERM OUTCOME AND QUALITY OF LIFE OF PATIENTS REQUIRING MULTIDISCIPLINARY INTENSIVE CARE UNIT ADMISSION AFTER CARDIACOPERATIONS Jean-Louis Trouillet, MD Andres Scheimberg, MD Albert Vuagnat, MD Jean-Yves Fagon, MD Jean Chastre, MD Claude Gibert, MD

Patients with organ failure or severe infection after cardiac operations may require prolonged stays in the intensive care unit. This study examined long-term mortality and determined quality of life for surviving patients in this group. This observational cohort study was conducted at Bichat Hospital, Paris, an academic tertiary care center. The study group consisted of 116 consecutive patients who underwent cardiac operations and were transferred to the multidisciplinary intensive care unit between January 1986 and December 1987. Patients referred for mediastinitis were automatically excluded. Respiratory failure (88.8%) and hemodynamic instability (81.9%) were the main causes of transfer; an infection was present in 23.3% of patients at entry into the intensive care unit. Twentyseven patients (23.3%) died in the intensive care unit. Presurgical New York Heart Association functional class, postoperative bacteremia before admission to the intensive care unit, and severity of illness on admission to the intensive care unit were independent predictors of death in the intensive care unit. After an average follow-up of 81 months (range 70 to 93 months), 69% of the patients alive at transfer from the intensive care unit were still alive. Preoperative New York Heart Association functional class was the only long-term independent prognostic factor. Quality of life, as evaluated by the Nottingham Health Profile, was good for more than 70% of the survivors and was not influenced by any recorded variables, with the exception of age. (J Thorac Cardiovasc Surg 1996;112:926-34)

onsiderable information has been published regarding operative death and long-term results after cardiac operations. 1'2 Moreover, several risk models have been developed to predict perioperative mortality and, more recently, perioperative morbidity. 3-14 Complications arising after heart operations remain frequent, however; three recent publications showed that 15% to 33% of the patients who survived cardiac operations had at least one such adverse event. T M Some of these complications are relatively easy to control, whereas others necessitate a long stay in an intensive care unit (ICU),

C

From the Service de Rdanimation M6dicale, H6pital Bichat, Paris, France. Received for publication Dec. 22, 1995; revisionsrequested Feb. 21, 1996; revisions received May 13, 1996; accepted for publication May 14, 1996. Address for reprints: Jean-Louis Trouillet, MD, Service de R6animation M6dicale, H6pital Bichat, 46, rue Henri Huchard, 75877 Paris Cedex 18, France. Copyright © 1996 by Mosby-YearBook, Inc. 0022-5223/96 $5.00 + 0 12/1/74904 926

with subsequent further morbidity and even mortality. Among these complications, mediastinitis is a distinct clinical entity that has a significant negative influence on long-term survival, independent of the patient's preoperative condition. 15 Respiratory failure, shock, hemodynamic instability, and multiorgan failure are among the most common indications for prolonged need for intensive care. The large number of patients transferred to our ICU for serious organ dysfunction or severe infection after cardiac operations provided us with the opportunity to prospectively construct, starting January 1986 and ending December 1987, a database with a number of clinical and biologic variables for this specific population. We decided to analyze the 2-year initial cohort of patients because no specific data were available concerning long-term follow-up. This study was examined long-term mortality and attempted to determine the impact of early complications of heart operations on quality of life for long-term survivors. The prospectively constructed database was used to determine whether late outcome could be predicted from clinical variables

The Journal of Thoracic and Cardiovascular Surgery Volume 112, Number 4

representing illness severity at the time of admission to the I C U .

Patients and methods Study population. Patients evaluated in this study were recruited from among those transferred because of severe postoperative complications between January 1986 and December 1987 from seven departments of cardiac surgery in Paris to our ICU. Of the 254 critically ill patients referred after operation during this period, 116 were enrolled in the study. Patients with acute mediastinitis (n = 108) or who had undergone cardiac transplantation (n = 1) were automatically excluded; 29 others were excluded because their usual places of residence were outside France. Data collection. Preadmission clinical characteristics of each patient included 13 variables: age, sex, New York Heart Association (NYHA) functional class 16 before cardiac operation, emergency versus elective operation, date of operation, type of operation, duration of cardiopulmonary bypass, use of intraaortic balloon pumping, use of inotropic agents (epinephrine, dobutamine, or dopamine > 5 /xgm. kg -1. rain -1) for longer than 24 hours, severe postoperative arrhythmia necessitating urgent treatment, reintervention (for tamponade, bleeding, coronary artery bypass grafting (CABG), or valvular dysfunction), cardiac arrest, and postoperative bacteremia. Other clinical characteristics were noted within the first 24 hours of admission to our ICU: number and type of organ dysfunctions or infection, defined according to the organ dysfunctions and infection number (ODIN) m o d e l t 7 ; s e ~ verity of illness, evaluated with the simplified acute physiology score (SAPS), is which is a scoring system similar to the Acute Physiology and Chronic Health Evaluation II (APACHE II; the correlation coefficient between these two scores was R 2 = 0.70 for cardiac surgical patients admitted to the ICU, unpublished data); Glasgow Coma Scale score; temperature; arterial oxygen tension and fraction of inspired oxygen; blood urea, creatinine, and bilirubin levels; transaminase and alkaline phosphatase activities; prothrombin time; platelet count; thoracic radiologic score~ determined according to a modified version of Weinberg's technique, with possible scores ranging from 0 to 1219; and use of inotropic agents at admission to the ICU. Finally, four other variables related to ICU stay were recorded: length of treatment with epinephrine and other inotropic agents (dobutamine or dopamine), duration of mechanical ventilation, hemodialysis, and length of stay in the ICU. The 116 patients were classified as nonsurvivors (died in the ICU) and survivors (discharged alive from the ICU). For nonsurvivors, cause of death was identified when possible. For survivors, follow-up interviews were conducted by telephone with the patients and their attending physicians between July 1993 and October 1993. For living patients, morbidity, NYHA class, and health status 2° were recorded during the last interview. In addition, quality of life for these patients was evaluated with part I of the French version of the Nottingham Health Profile. z~ This part contains 38 statements covering feelings and functions in six areas: pain, energy, physical

TrouilIet et aL

927

mobility, sleep, social isolation, and emotional reactions. The responses to the statements are scored on a scale of 0 to 100 for each category, with lower scores indicating fewer difficulties. Patients' levels were compared with those obtained for community-based age- and sexmatched control subjects with no adverse health conditions (or minor, nonacute conditions). Subjects were then divided into two classes, younger than 55 years and 55 years old or older, according to stratification used by Leplege (personal communication). 2I Long-term survivors were then assigned to one of two groups. Group I consisted of patients with at least two items of the Nottingham Health Profile part I greater than the mean score plus 1 standard deviation (SD); they were considered to have a deteriorated condition or a poorer quality of life than expected. Group II consisted of patients whose condition was stable or better than expected. The patients were also asked a number of other pertinent questions, including those regarding return to work and recollection of their previous ICU stay. Statistical analysis. Data, reported as means (_+ SD), were analyzed with the SAS software package (Statistical Analysis System, Cary, N.C.). The ~ test and Fisher's Exact Test were used to assess differences among dichotomous variables. The unpaired Student's t test was used to analyze continuous variables for significant differences. For variables that were not normally distributed, the Mann-Whitney U test was used. Variables with univariate p value less than 0.05 were then used in a logistic regression model to predict the occurrence of death in the ICU and by the end of follow-up. A similar analysis was made to predict quality of life. Univariate logistic regression was also used for variables withp value less than 0A0. Survival curves were calculated by the Kaplan-Meier method and compared with standard log-rank tests.

Results Patient profiles and short-term outcome. Preadmission clinical characteristics and data r e c o r d e d within 24 hours of admission to the I C U of the 116 patients are shown in Tables I and II. Approximately 40% of all heart operations were for C A B G . Valve operation alone was p e r f o r m e d in 38% of cases; c o m b i n e d procedures were p e r f o r m e d in 10 cases. The other procedures were aortic operations in six cases (to repair dissection or thoracic aneurysm), left ventricular a n e u r y s m e c t o m y in five cases (associated with valve replacement in one case and closure of a ventricular septal defect in another), repair of atrial or ventricular septa] defect or m o r e complex congenital heart disease in five cases, and p e r i c a r d e c t o m y in one case. T h e median interval between the surgical p r o c e d u r e and the transfer to our I C U was 5 days. Respiratory failure, as defined in the O D I N m o d e l 17 (arterial oxygen tension < 6 0 m m H g with 0.21 fraction of inspired oxygen or the n e e d for ventilatory support), and shock or h e m o -

928

The Journal of Thoracic and Cardiovascular Surgery October 1996

Trouillet et al.

Table I. Preadmission clinical characteristics of study population Survivors (n = 89) Variable

No. (or value)

Age (yr) S e x (M/F) Preoperative N Y H A functional class I II III IV Emergency operation Duration of CPB (min) Intraaortic balloon pumping Surgical procedure CABG Valve operation Valve operation + C A B G Miscellaneous Initial postoperative complications Severe arrhythmia Cardiac arrest Reintervention for tamponade, bleeding, or dysfunction Postoperative use >24 hr of inotropic agents Interval between operation and admission to I C U (days) Postoperative bacteremia

60.5 + 12.2 55/34

%

Nonsurvivors (n = 27) No. (or value)

%

64.6 _+ 11 21/6

27 18 34 10 11 101 _+ 39 25

30.3 20.2 38.2 11.2 12.4

0.1 0.16 0.004 3.8 34.6 23 38.5 33.3

28.i

1 9 6 10 9 114 _+ 50 10

40 35 6 8

44.9 39.3 6.7 9

5 9 4 9

18.5 33.3 14.8 33.3

24 9 12 75 6 -+ 3 11

27 10.1 13.5 84.3

11 5 5 24 8.2 -+ 7.5 12

40.7 18.5 18.5 88.9

12.4

p

37

44.4

0.02 0.2 0.4 0.003

0.2 0.4 0.5 0.8 0.1 0.0007

Each patient's characteristics were recorded on a standard data chart at admission to ICU, with the exception of NYHA (n = 115, one nonsurvivor is missing) and bypass duration (n - 106). CPB, Cardiopulmonary bypass.

dynamic instability (systolic arterial pressure <90 mm Hg with signs of peripheral hypoperfusion or continuous infusion of vasopressor or inotropic agents required to maintain systolic pressure >90 mm Hg) were reasons for referral in 88.8% and 81.9% of cases, respectively. Infection (clinical evidence of infection associated with positive results of blood cultures or the presence of gross pus in a closed space) was present in 23.3% of the patients at transfer to the ICU. It is important to notice the predominance of patients with multiple organ dysfunction: one hundred patients (86.2%) had three or more organ dysfunctions, infection of any type, or both. The mean SAPS was 11.2 points, corresponding to an APACHE II score of 17.6, with a mean calculated APACHE II score of 15.7 for the survivors and 24.1 for the nonsurvivors. Mean duration of stay was 16.3 _+ 20.3 days for survivors and 22.7 _+ 32.9 days for nonsurvivors. The overall mortality rate in the ICU was 23.3%. Causes of death were neurologic problems in five cases, cardiogenic shock in six, septic shock in nine, and miscellaneous causes in three; the other four patients who died had clinical pictures of febrile shock with multiple organ failure of uncertain or complex origin. Of the 34 variables listed in Tables I and II,

18 were found to be significantly associated with death in the ICU. Logistic regression analysis identified SAPS on the day of admission to the ICU (p < 0.0001), presurgical NYHA functional class (p = 0.004), and postoperative bacteremia before admission (p < 0.01) as independent predictors of death. The study population was further divided into two subgroups according to the lag time before transfer to our ICU, with a cutoff of 5 days. Application of the multivariate model to each subgroup showed SAPS to be a unique independent predictor of death in the ICU. Long-term outcome. Of the 89 patients discharged from the ICU, long-term follow-up was obtained for 86. Eight patients died during hospitalization but after discharge from the ICU, with a mean survival from the ICU of 13.6 days. All deaths except two (one of mediastinitis that appeared more than 10 days after discharge from the ICU and one of stroke) were attributed to the severity of the cardiopathy. For the 19 other patients who died after discharge from the hospital, median survival after discharge from the ICU was 48 months (range 7 to 81 months). Deaths in this group resulted directly from the cardiopathy in eight cases (intractable heart failure in five cases, acute myocardial

The Journal of Thoracic and Cardiovascular Surgery Volume 112, Number 4

Trouillet et al.

929

Table II. Characteristics of 116 patients on day of admission to ICU I C U Survivors (n = 89)

I C U Non Smvivors (n = 27)

Variable

No. (or value)

%

No. (or value)

%

p

Organ dysfunction (%) Cardiovascular Respiratory Renal Neurologic Hepatic Hematologic Infection Number of dysfunctional organs and/or infection

70 78 14 11 8 1 17

78.7 87.6 15.7 12.4 9 1.1 19.1

25 25 14 13 7 3 10

92.6 92.6 51.9 48.1 25.9 11.1 37.0

0.15 0.7 0.0005 0.0002 0.04 0.04 0.07 0.0001

14 51 24 9.7 _+ 4 13.1 + 3 37.9 _+ 0.8 210.32 ± 104.75 16.9 _+ l l 185 ± 152 46 ± 38 97 ± 195 i37 ± 103 68 + 17 201 + 150 5.7 ± 3.7 69

15.7 57.3 27

2 4 21 16.2 -+ 6 9.5 -+5 38.1 ± 1.1 187.59 -+ 79.1 26.2 _+ 14.6 339 -+ 235 78 -+ 62 779 -+ 1625 149 ± 104 50 ± 21 130 ± 104 8.2 + 3.3 25

7.4 14.8 77.8

<3 3 >3 SAPS* GCS score*

Temperature (°C)* Pao2/Fio2*

Blood urea (retool/L)* Creatinine (/xmol/L)* Bilirubin (mmol/L)* A S T (IU)*

Alkaline phosphatase* Prothrombin index (%)*? Platelet count (103 cells/ram3) * Radiologic score* Inotropic agents (%)

77.5

92.6

0.0001 0.0001 0.4 0.25 0.0006 0.0001 0.001 0.0002 0.6 0.0001 0.02 0.0002 0.09

GCS, Glasgow Coma Scale; Pao 2, arterial oxygen tension; Fio2, inspired oxygen fraction, AST, aspartate aminotransferase. *Mean _+ SD.

]-None had received oral anticoagulants.

infarction in two, and sudden death in one), from prosthetic valve endocarditis in one case, and from miscellaneous causes in six cases (cancer in three cases, cirrhosis in one, after laryngeal operation in one, and hemorrhagic complication of anticoagulant treatment in one); the cause of death could not be assessed in the remaining four cases because of inadequate records. At a median follow-up of 81 months (range 70 to 93 months), 59 of the 86 patients (68.6%) were still alive. Analysis of clinical data on these 59 patients (mean age at end of follow-up 65.6 years, range 25 to 85 years; 39 male patients) showed that surgical interventions were distributed as follows: 25 CABG (42.4%), 26 valve replacements (44.1%), five combined procedures (8.5%), and three other operations (5.1%). Analysis of the 34 parameters recorded at the ICU admission and four additional clinical data related to ICU stay (length of treatment with epinephrine and other inotropic agents, duration of mechanical ventilation, hemodialysis, length of stay

in ICU), with respect to patients who died after ICU discharge (n = 27) and long-term survivors (n = 59) identified only three statistically different factors in univariate analysis: preoperative NYHA functional class (p < 0.005), duration of epinephrine therapy during ICU stay (p = 0.03), and emergency operation (p < 0.05). Results of univariate logistic analysis are shown in Table III. By multiple logistic regression analysis, the only variable predictive of postponed death was NYHA functional class, with an odds ratio of 12 for class IV versus other classes (confidence interval 2.3 to 61.5, p = 0.003). Fig. 1 shows the Kaplan-Meier survival analysis of the 86 ICU survivors stratified according to preoperative NYHA class. Observed 6-year survival varied from 86% for class l to 20% for class IV. The survival curve depicted for class IV differs significantly from those for the other classes (log-rank test, p < 0.005). Quality of life for long-term survivors. NYHA functional class and chronic health status at the time of last interview (October 93) are shown in Fig. 2. Eighty percent of long-term survivors were in

930

The Journal of Thoracic and Cardiovascular Surgery October 1996

Trouillet et al.

I

t m I

0.8

@

0.6

I

a.

II III

T 1

0.4 ,.Q

2

IV

0.2

0

I

l

l

l

l

0

20

40

60

80

T i m e after I C U Discharge (months) Fig. 1. Kaplan-Meier survival analysisof the 86 survivors grouped by preoperative NYHA functional class. The p value of the overall log-rank test comparing the survival distribution in the four NYHA classes is 0.002.

Table III. Comparison with univariate logistic regression of clinical parameters* between patients who died after discharge from the ICU (n = 27) and long-term survivors (n = 59) Parameter estimate Preoperative NYHA functional class IV Emergency operation Duration of CPB (min) Use of inotropic agents >24 hr during postoperative period Duration of epinephrine therapy during ICU stay (days)

2.48 1.36 0.012 1.89 0.26

OR 12.00 3.93 1.01 6.64 1.30

95% Cl for OR

p

2.34 - 61.5 1.01-15.3 0.99-1.02 0.82-54.0 0.98 -1.73

0.003 0.049 0.082 0.077 0.069

OR,Odds ratio; C/, confidenceinterval; CPB, cardiopulmonarybypass. *Comparison of the 33 other variables did not show differencessignificantat p ~< 0.10.

NYHA functional class I or II, and 75% were at health status level A or B (the less severe chronic organ system insufficiency states). During follow-up, 26 cardiovascular complications were noted in 23 patients; three of these patients had two complications. The complications were angina pectoris (nine), valve dysfunction (five, all reoperated), episodes of congestive heart failure (four), arrhythmias or conduction disturbances (four), and embolic events (four). Quality of life was also evaluated for 54 of the 59 long-term survivors by means of the Nottingham Health Profile questionnaire part I. Data could not be obtained for five patients because these patients were unable to complete the interview for the following reasons: three had neurologic disease, one was not a French speaker, and one resided far from

France at the intended date of interview. Fig. 3 illustrates the mean scores for each of the six areas evaluated for our patients compared with those obtained for a community-based control population. Among all the comparisons made, only physical mobility for patients 55 years old or older was statistically significantly different from the value for the general population (Fig. 3, B). Moreover, it must be noted that 16 patients had two items with scores greater than the mean plus 1 SD, but only four patients had two items with scores greater than the mean plus 2 SDs. Additional questions were put to patients. Of the 26 who worked for a living before the cardiac operation, 13 had returned to work after discharge from the hospital, 12 patients decided to retire, and one was unemployed. None of the 48 patients who

The Journal of Thoracic and Cardiovascular Surgery Volume 112, Number 4

Trouillet et al.

40

40

351

3530-

30~

25-

25-

20-

2o15-

15Z

Z 10-

105r/i///IA

0 I A

931

II

llI

0 -

IV

NYHA Class

A B

B

2

C

D

Health Status

Fig. 2. NYHA functional class (A) and health status (B) of the 59 tong-term survivors at the last follow-up interview. A, Good health, no functional limitations; B, mild to moderate limitations of activity because of chronic medical problem; C, chronic disease producing serious but not incapacitating restriction of activity. D, severe restriction of activity because of disease, includes persons bedridden or institutionalized because of illness. received mechanical ventilation had bad memories of the ICU stay (all patients were sedated during mechanical ventilation). Eight patients reported dysphonia, however, probably as a result of prolonged intubation, and one reported slight cutaneous pain at the site of healed pressure ulcers. In an attempt to predict the quality of life 6 years after discharge from the ICU, the 54 long-term survivors were divided into two groups (see Patients and Methods section for details). Group I (worse than expected) included 16 patients, group II (stable or better than expected) comprised remaining patients. Results of univariate logistic analysis of the 38 variables quoted in the data collection section are shown in Table IV. The only variable associated with the quality of life was age. Mean age at entry of the study was 64.4 _+ 8.6 years in group I versus 57.4 + 12.7 years in group II (p < 0.05). Discussion

In recent years, cardiovascular procedures have become the most commonly performed major operations in industrial countries. Surgical indications have become broader for older patients, and now patients with more severe conditions undergo operation. Undoubtedly, these patients require more intensive care resources and need more detailed explanations before they give their informed consent. Mortality and morbidity after cardiac operations are both important indicators of quality of

care. Mortality alone cannot be the sole marker because the patients who benefit the most from operation are often paradoxically those at higher risk. We know that the cost associated with operation is highly influenced by the complexity of complications and duration of ICU stay, but we have little information about ultimate impact of these factors on the patient's subsequent quality of life. The mare results of this study demonstrate that long-term survival was quite impressive when a patient's status required prolonged ICU stay after a cardiac operation, with a fair-to-good quality of life despite a notable initial mortality rate. Preoperative cardiac condition, as established by NYHA functional class, was a major determinant of patient survival, and only age had an influence on long-term quality of life after hospital discharge. Nevertheless, it is important to note that additional factors may affect survival but did not achieve significance because of the relatively small number of patients in the study group. Moreover, the factors significant in the univariate models but not in the multivariate model may have been significant only because of confounding by the significant variables in the multivariate model. Points worth noting in this study are that our patients were recruited by referral to our center by several surgical teams, rather than by prospective evaluation of patients at one institution requiring prolonged intensive care. For this reason, the study

932

The Journal of Thoracic and Cardiovascular Surgery October 1996

Trouillet et al.

30I[

25-

I

30 ] 25

[~ Studypopulation [] Normalpopulation

20-

I

[~ Studypopulation

I

&

20 432

15

y427

424

~

44i

~

0

0

Mobility Sleep Energy Pain A

Social Emotional

0 Mobility Sleep Energy Pain

Age<55 yr

B

Social Emotional

Age->55 y r

Fig. 3. Nottingham Health Profile part I at last follow-up interview. Comparison of mean scores obtained for study population and control community population, divided according to age: younger than 55 years (A) and 55 years old or older (B). Number above each column refers to number of persons interviewed.For each category, possible scores ranged from 0 to 100; lower scores denote fewer difficulties.Asterisk indicates p < 0.001.

Table IV. Comparison with univariate logistic regression of cIinical parameters* in 54 long-term survivors according to adjusted health functional status (group I vs group II) Parameter estimate

OR

95% CI for OR

p

0.063 2.144 1.810

1.065 8.538 6.111

0.998-1.137 0.815-89.51 0.717-52.06

0.058 0.074 0.098

A g e (yr) Emergency operation Cardiac dysfunction on the day of admission to I C U

OR, Odds ratio; CI, confidence interval. '~Comparison of the 33 other variables did not show differences significant atp ~ 0.10.

population may not be representative of patients with major complications after cardiac operations. Patients referred to our ICU were estimated to represent 4% of the total surgical volume during the study period. We observed a mortality rate of 23% in the ICU and an in-hospital mortality rate of 33%. A review of other studies that evaluated patient mortality in cases of prolonged stay in the ICU after cardiac operations demonstrated remarkably similar findings. Overall, other investigators reported rates in the range of 20% to 30%. For instance, a report from the Department of Veteran's Affairs Cooperative Study9 gave mortality rates of 24% and 25% for patients undergoing CABG and valve-replacement, respectively, when mechanical ventilation was required for longer than 48 hours. A Medicare patien t survey showed that for patients 65 years old or older the frequency of death within 30 days of admission was 24.4% when mechanical ventilation

lasted longer than 48 hours. 11 In a study reporting the experience of three London hospitals, patients who were in the ICU for more than 48 hours had a mortality rate in the ICU of 29% and an in-hospital mortality rate of 33.3%. 22 Finally, in a private teaching hospital, Kollef, Wragge, and Pasque ~3 observed a 19.6% mortality rate in the ICU for patients requiring mechanical ventilation for longer than 48 hours. In light of the demographic details and mortality rates in those studies, our ICU population appears to be a reasonable reflection of patients undergoing cardiac operations and requiring prolonged intensive care. It should be recalled that mediastinitis and heart transplantation, which represent distinct clinical entities, were chosen as exclusion criteria to enhance the homogeneity of the patient population. Analysis of long-term outcome showed that survival during the first year after leaving the ICU was

The Journal of Thoracic and Cardiovascular Surgery Volume 112, Number 4

89.5%; survival was as high as 98.7% when only patients leaving the hospital alive were considered. During the first 5 years, corresponding rates were 71% and 78%. These results seem better that those observed for patients admitted to nonspecialized ICUs. For instance Ridley and colleagues 24 observed that of 378 patients discharged from the ICU, 68% (compared with 81.4% in our series) were still alive at 2 years after discharge from the ICU. Similarly, Stauffer and associates 25 noted a 1-year survival of 58% for patients after discharge from the hospital. Results were even worse for elderly patients 26 or patients with chronic obstructive pulmonary disease who required mechanical ventilation for acute respiratory failure. 27' 2s Large studies on cardiac surgical patients showed a 75% survival at 5 years for patients after leaving the hospital after heart valve replacement, 1 and 92% survival rate at 5 years for patients who underwent CABG. 2 It is encouraging to note that long-term survival is only slightly worse for cardiac surgical patients who survived to be discharged from the ICU. Our results are in accordance with a report on British patients who had 1-year survivals of 85% and 90.7% for patients who survived to be discharged from the I C U and the hospital, respectively. 22 Unlike previous studies, which concluded that, age and severity of illness were long-term independent prognostic factors in a general population, 25 the only variable that emerged from our series was advanced disability before operation, as assessed by the N Y H A functional classification. Even in univariate analysis, age, SAPS, and the number of organs involved in acute failure did not differ according to long-term survival for patients who survived to be discharged from the ICU. The quality of life and health status level of surviving patients 6 years after a prolonged stay in I C U were good. Increasing efforts have been made to study this important phase for patients with chronic diseases, but to the best of our knowledge no information is available on the long-term prognosis of those who survived acute complications of cardiac operations but left the hospital. W e chose to use a rather cumbersome but well-established questionnaire; the Nottingham Health Profile has previously been used after C A B G 29 and is the only score for which a French translation has been validated. 2~ Nottingham Health Profile (part I) showed here that, other than a slightly higher score in the degree of physical mobility for older patients, the scores seemed similar to those of the general French

Trouillet et al.

933

population. Moreover, measurements of quality of life showed favorable or stable outcomes for 70% of the long-term survivors. Similarly, N Y H A classification and chronic health status provided evidence of a clear benefit for a high percentage of these patients. Despite the high early mortality rate, intensive care is therefore justified for patients with acute life-threatening complications of cardiac operations, unless it is clear that there is no prospect of recovery from the acute episode. We are indebted to A. Leptege, INSERM U292, Mesure de la Sant6 Perceptuelle etde la Qualit6 de Vie, H6pital de Bic~tre, Le Kremlin Bic~tre, France, who provided the data concerning the levels of Nottingham Health Profile obtained for community-based age and sex norms in a general population. We thank C. Brun and A. Failin for assistance in the preparation of the manuscript. REFERENCES 1. Hammermeister KE, Sethi GK, Henderson WG, Oprian C, Kim T, Rahimtoola S. A comparison of outcomes in men 11 years after heart-valve replacement with a mechanical valve or bioprosthesis. Veterans' Affairs Cooperative Study on Valvular Heart Disease. N Engl J Med 1993;328:1289-96. 2. Yusuf S, Zucker D, Peduzzi P, Fisher LD, Takaro T, Kennedy JW, et al. Effect of coronary artery bypass graft surgery on surv!vai:overviewof 10-yearresults from randomised trials by the Coronary Artery Bypass Graft Surgery Trialists Collaboration. Lancet 1994;344:563-70. 3. Parsonnet V, Dean D, Bernstein AD. A method of uniform stratification of risk for evaluating the results of surgery in acquired adult heart disease. Circulation 1989;79(6 Pt 2):I312. 4. Weintraub WS, Jones EL, Craver J, Guyton R, Cohen C. Determinants of prolonged hospital stay after coronary bypass surgery. Circulation 1989;80:276-84. 5. Hannan EL, Kilburn H Jr, O'Donnel JF, Lnkacik G, Shields EP. Adult open heart surgery in New York state: an analysis of risk factors and hospital mortality rates. JAMA 1990;264: 2768-74. 6. Hannan EL, Kilburn H, Racz M, Shields E, Chassin MR. Improving the outcomes of coronary artery bypass surgery in New York state. JAMA 1994;271:761-6. 7. O'Connor GT, Plume SK, Olmstead EM, Colfin LH, Morton JR, Maloney CT, et al. Multivariate prediction of in-hospital mortality associated with coronary artery bypass graft surgery. Northern New England Cardiovascular Disease Study Group. Circulation 1992;85:2110-8. 8. Edwards FH, Clark RE, Schwartz M. Coronary artery bypass grafting: the Society of Thoracic Surgeons national database experience. Ann Thorac Surg 1994;57:12-9. 9. Hammermeister KE, Burchfiel C, Johnson R, Grover FL. Identification of patients at greatest risk for developingmajor complications at cardiac surgery. Circulation 1990;82(5 Suppl):IV380-9. 10. Tnman KJ, McCarthy RJ, March RJ, Najafi H, Ivankovich AD. Morbidity and duration of ICU stay after cardiac surgery. Chest 1992;102:36-44. 1i. Geraci JM, Rosen AK, Ash AS, McNiffKJ, Moskowitz MA.

934

12.

13.

14.

15.

16.

17.

18.

19.

The Journal of Thoracic and Cardiovascular Surgery October 1996

Trouillet et al.

Predicting the occurrence of adverse events after coronary artery bypass surgery. Ann Intern Med 1993;118:18-24. Higgins TL, Estrafanous FG, Loop FD, Beck GJ, Blum JM, Parandi L. Stratification of morbidity and mortality outcome by preoperative risk factors in coronary artery bypass patients: a clinical severity score. JAMA 1992;267:2344-8. Tu JV, Jaglal SB, Naylor CD, Steering Committee of the Provincial Adult Care Network of Ontario. Multicenter validation of a risk index for mortality, ICU stay, and overall hospital length of stay following cardiac surgery [published erratum appears in Circulation 1991;84:446]. Circulation 1995;91:677-84. Becker RB, Zimmerman JE, Knaus WA, Wagner DP, Seneff MG, Draper EA, et al. The use of APACHE II1 to evaluate ICU length of stay, resource use, and mortality after coronary artery by-pass surgery. J Cardiovasc Surg (Torino) 1995;36: 1-11. Milano CA, Kesler K, Archibald N, Sexton D J, Jones RH. Mediastinitis after coronary artery bypass graft surgery: risk factors and long-term survival. Circulation 1995;92:2245-51. The Criteria Committee of the New York Heart Association. Nomenclature and criteria of diagnosis of diseases of the heart and great vessels. 8th ed. Boston: Little, Brown, 1979. Fagon JY, Chastre J, Novara A, Medioni P, Gibert C. Characterization of intensive care unit patients using a model based on the presence or absence of organ dysfunctions and/or infection: the ODIN model. Intensive Care Med 1993;19:137-44. Le Gall JR, Loirat P, Alperovitch A, Glaser P, Granthil C, Mathieu D, et al. A simplified acute physiology score for ICU patients. Crit Care Med 1984;12:975-7. Fagon JY, Chastre J, Hance AJ, Guiguet M, Trouillet JL, Domart Y, et al. Detection of nosocomial lung infection in ventilated patients: use of a protected specimen brush and

20.

21.

22.

23.

24. 25.

26.

27.

28.

29.

quantitative culture techniques in 147 patients. Am Rev Respir Dis 1988;138:110-6. Knauss WA, Zimmerman JE, Wagner DP, Draper EA, Lawrence DE. APACHE: acute physiology and chronic health evaluation: a physiologically based classification system. Crit Care Med 1981;9:591-7. Bucquet D, Condon S, Ritchie K. The French version of the Nottingham Health Profile: a comparison of items weights with those of the source version. Soc Sci Med 1990;30:829-35. Holmes L, Loughead K, Treasure T, Gallivan S. Which patients will not benefit from further intensive care after cardiac surgery? Lancet 1994;344:1200-2. Kollef MH, Wragge T, Pasque C. Determinants of mortality and multiorgan dysfunction in cardiac surgery patients requiring prolonged mechanical ventilation. Chest 1995;107: 1395-401. Ridley S, Jackson R, Findlay J, Wallace P. Long-term survival after intensive care. BMJ 1990;301:1127-30. Stauffer JL, Fayter NA, Graves B, Crumb M, Lynch JC, Goebel P. Survival following mechanical ventilation for acute respiratory failure in adult men. Chest 1993;104:1222-9. Chelluri L, Pinsky MR, Donahoe MP, Grenvik A. Long-term outcome of critically ill elderly patients requiring intensive care. JAMA 1993;269:3119-23. Menzies R, Gibbons W, Goldberg P. Determinants of weaning and survival among patients with COPD who require mechanical ventilation for acute respiratory failure. Chest 1989;95:398-405. Hudson LD. Survival data in patients with acute and chronic lung disease requiring mechanical ventilation. Am Rev Respir Dis 1989;140(2 Pt 2):$19-24. Caine N, Harrison SCW, Sharpies LD, Wallwork J. Prospective study of quality of life before and after coronary artery bypass grafting. BMJ 1991;302:51!-6.