Hospital Outcomes for Patients with Stage III and IV Lung Cancer Admitted to the Intensive Care Unit for Sepsis-Related Acute Respiratory Failure

Abstract

Background:

In recent years, intensive care for cancer patients has improved and treatment of critically ill cancer patients has become increasingly aggressive over time. However, not all cancer patients would benefit from aggressive care, especially those with late-stage cancer.

Objective:

We aimed to investigate the outcome of late-stage lung cancer patients with sepsis-related respiratory failure and identify predictors of mortality.

Methods:

From 2007 to 2008, consecutive stage III and IV lung cancer patients admitted to an intensive care unit (ICU) of a teritiary medical center in Taiwan for sepsis-related respiratory failure were retrospectively enrolled. Data at baseline and upon ICU admission were collected. In-hospital survival was analyzed. Variables of the survivors to hospital discharge and patients who died were compared by uni- and multivariate analyses.

Results:

Seventy patients were enrolled. During a mean follow-up period of 30.10 days, 29 (41.4%) patients survived to hospital discharge and 41(58.6%) died. Compared with the survivors, the patients who died had poor performance status, lower serum albumin level, higher percentage of disseminated intravascular coagulation, and more severe organ dysfunction as disclosed by higher Sequential Organ Failure Assessment (SOFA) scores. Multivariate analyses revealed that SOFA score (p=0.026) was the only independent predictor of mortality; 44.8 % (13/29) of survivors were weaned from ventilator during hospitalization.

Conclusion:

Among late-stage lung cancer patients with sepsis-related respiratory failure, those with lower SOFA scores seemed to have better survival rate and may benefit from intensive care in the ICU. Early palliative care should be considered for all patients with advanced lung cancer, and hospice care is suggested for those with sepsis-respiratory failure and high SOFA scores.

Introduction

Lung cancer is the leading cause of cancer death worldwide.¹ With advances in chemotherapeutic agents, patients with advanced lung cancer have longer disease control and survival nowadays, but almost all of them will eventually die of cancer progression or cancer-associated complications.² When catastrophic complications emerge, conflicts between aggressive treatment and hospice care are often hard to reconcile for patients and their families, and even for their physicians.

In cases of life-threatening events, such as sepsis, cancer patients used to be deemed to have a poorer prognosis than noncancer patients.^3–5 In the past decade, intensive care for cancer patients has improved so much that their intensive care unit (ICU) outcome approximates that of noncancer populations.⁶ A recent report revealed that the survival of nonsurgical patients with lung cancer requiring ICU admission was 37% after 90 days.⁷ Inspired by more and more optimistic evidence, the treatment of critically ill cancer patients has become increasingly aggressive over time.^7–9 Although much effort has been made to study predictors for better outcomes, determining the optimal time to shift aggressive treatment to hospice care is still an open question, especially for patients facing critical illness, such as sepsis-related respiratory failure, which was previously a common reason for such patients to be admitted to ICU.^6,10,11 Investigation specific to this subpopulation may help clinicians counsel patients and families about this decision

In this study, we retrospectively analyzed patients with late-stage lung cancer, who were admitted to our ICU in Taipei Veterans General Hospital for mechanical ventilator support for sepsis–related respiratory failure from 2007 to 2008, and strived to identify the predictors of mortality, which may facilitate making difficult decisions about switching from aggressive treatment to hospice care.

Methods

Study subjects

This study was conducted at Taipei Veterans General Hospital, a tertiary referral medical center in Taiwan. The study protocol was approved by the Institutional Review Board of Taipei Veterans General Hospital. During the period from January 2007 to December 2008, consecutive patients with stage III to stage IV lung cancer, who were admitted to ICU for sepsis-related respiratory failure, were enrolled for study.

Data

Data of study subjects included demographics, baseline condition (Eastern Cooperative Oncology Group performance status [ECOG-PS]) at or within 1 month prior to ICU admission, baseline spirometry within 3 months prior to ICU admission, laboratory profiles (complete blood count/differential count, biochemistry tests, arterial blood gas) and clinical assessment upon ICU admission (Glasgow Coma Scale [GCS] scores, Acute Physiology And Chronic Health Evaluation II [APACHE II] scores, Sequential Organ Failure Assessment [SOFA] scores, presence of disseminated intravascular coagulation [DIC], shock, acute respiratory distress syndrome [ARDS], need for hemodialysis). The blood tests were performed at our central laboratory by an automated hematology analyzer (LH750, Beckman Coulter), automatic chemistry analyzer (Hitachi modular 7600, Hitachi Ltd.) and blood gas anlyzer (ABL5, Raiometer).

The ECOG-PS score is a measure of a cancer patient's general well-being and activities of daily life, which varies from 0 to 5 with 0 being asymptomatic and 5 being dead.¹² The SOFA score, initially named Sepsis-Related Organ Failure Assessment score,¹³ is a scoring system to assess the extent of a patient's organ function or rate of failure, which are composed of scores from six organ systems, including respiratory, cardiovascular, hepatic, coagulation, renal, and neurological systems, with each graded from 0 to 4 according to the degree of dysfunction/failure.¹⁴ The higher the SOFA scores, the more severe the overall degree of organ dysfunction.

Follow-up and outcome measures

The study subjects were followed up from enrollment to hospital discharge or until death during that admission. Patients were discharged home by their physicians following stabilization and recovery of sepsis and other treatable complications. If patients failed to be weaned from the ventilator, they were transferred to a respiratory care facility for long-term ventilator support. A patient might also be discharged with a critical condition and against their physician's advice. However, that patient's death could be confirmed later when the family came back to apply for documents necessary for obtaining a death certificate. The survival rate in different time frames (7, 14, and 30 days after enrollment) and at hospital discharge (in-hospital survival) was computed. At hospital discharge, the status of ventilator use was also recorded to compute the weaning rate.

Statistical analysis

Statistical analysis was performed utilizing SPSS (Version 15.0, SPSS Inc., Chicago, IL) and Stata 11 (StataCorp. 2009. Stata Statistical Software: Release 11. College Station, TX: StataCorp LP) software. All data were expressed as mean±standard deviation (SD) or percentage. Comparisons between two groups were determined by independent Student's t test for continuous variables or Pearson's χ² test/Fisher's exact test for categorical variables as appropriate. Variables found significant at p<0.05 in a univariate analysis were subjected to binary logistic regression for multivariate adjustment. The method of entering variables was “ENTER” in the SPSS software. Survival analysis was assessed using a Kaplan–Meier analysis, with the significance based on the log-rank test. Statistical significance was inferred at a two-sided p value of<0.05.

Results

During the period from January 2007 to December 2008, 70 patients were recruited for study. These were consecutive patients, and all patients with stage III or IV lung cancer were included. The enrollees were of pure Chinese ethnicity, older in age (mean age: 74.43±12.02 years), the group was predominantly composed of male (60/70=85.7%), late-stage (stage IIIb-IV: 67/70=95.7%) patients, and all were intubated with support of mechanical ventilation. During a follow-up period of 30.10±24.81 days, 29 (41.4%) patients survived to hospital discharge and 41(58.6%) died during hospitalization (Table 1). The survivors to hospital discharge had better baseline performance status (ECOG-PS 3-4: 39.4% vs. 64.3%, p=0.032) as compared with the patients who died. Additionally, the survivor group had higher forced expiratory volume in one second/forced vital capacity (FEV₁/FVC) ratio (75.7±14.3 vs. 66.4±12.0, p=0.053, and more stroke patients (10.3% vs. 0, p=0.067), but neither reached statistical significance. Among the survivors to hospital discharge, 44.8% (13/29) were successfully weaned from ventilator support and discharged home. The rest (16/29) were ventilator dependent and were transferred to a chronic respiratory care facility.

Table 1.

Baseline Data of Study Subjects

	Survived to hospital discharge	Died	p
Number	29	41
Age	71.7±12.3	76.4±11.6	0.105
Male	23 (79.3%)	37 (90.2%)	0.299^a
Histology of lung cancer
Adenocarcinoma	11 (37.9%)	25 (61%)	0.097
Non-adenocarcinoma	18 (62.1%)	16 (39%)
Stage
IIIa	1 (3.4%)	2 (4.9%)	1^a
IIIb-IV (advanced)	28 (96.6%)	39 (95.1%)
Pulmonary function
FEV1/FVC (%)	75.8±14.3	66.4±12.0	0.053
FEV1 (% predicated)	73.7±24.0	77.0±18.8	0.670
ECOG performance status
0–2	18 (62.1%)	12 (29.3%)	0.013
3–4	11 (37.9%)	29 (70.7%)
Underlying comorbidities
Diabetes mellitus	7 (24.1%)	6 (14.6%)	0.487
Heart failure	9 (31%)	6 (14.6%)	0.177
Stroke	3 (10.3%)	0	0.067^a
COPD	6 (20.8%)	13 (31.7%)	0.454

Data are given as mean±SD or percentage. P values for comparisons between two groups are determined by Student's t test for continuous variables or χ² test /Fisher's exact test^a for categorical variables.

FEV1, forced expiratory volume in one second; FVC, forced vital capacity; ICU, intensive care unit; ECOG, Eastern Cooperative Oncology Group; COPD, chronic obstructive pulmonary disease.

Comparing the data upon ICU admission between the two studied groups, there were statistically significant differences in terms of SOFA scores (5.9±2.0 vs. 7.9±3.5, p=0.003), percentage of DIC (0 vs. 17.1%, p=0.036), and level of albumin (2.6±0.5 vs. 2.2±0.5, p=0.023), but there were no differences in APACHE II scores, GCS scores, origin of sepsis, PaO2/FiO2 ratio, number with renal failure receiving hemodialysis, shock, ARDS, blood cell counts, and all biochemistry profiles, except serum albumin level (Table 2). Most of the origin of sepsis was pulmonary infection, except one instance from a catheter-related infection. Approximately 50 % or more of study subjects were hospitalized related to treatment with either chemo- or radiotherapy (Table 2). Multivariate analysis incorporating baseline performance status, SOFA scores, DIC, and albumin level revealed that the SOFA score (OR: 1.360, 95% CI: 1.038–1.782, p=0.026) was the sole predictor of mortality (Table 3). If we dichotomized the SOFA scores of our subjects into two groups (6≦versus ≥7), the patients with SOFA scores ≥7 experienced a 6.148-fold risk for mortality (95% CI: 1.545–24.465, p=0.010).

Table 2.

Patients' Data on ICU Admission

	Survived to hospital discharge	Died	p
Mechanical ventilation	29 (100%)	41 (100%)
Timing of sepsis attack			0.865
Concurrent with diagnosis of cancer	3 (10.3%)	7 (17.1%)
During treatment for cancer	17 (58.6%)	21 (51.2%)
Rapid deterioration unrelated to cancer therapy	9 (31%)	12 (31.7%)
ICU stay	19.6±11.0	20.1±14.0	0.867
APACHE II scores	23.5±4.6	24.9±7.8	0.363
SOFA scores	5.9±2.0	7.9±3.5	0.003
Glasgow coma scale scores	10.1±1.4	9.4±2.4	0.187
Origin of sepsis
pulmonary	29 (100 %)	40 (97.6%)	1^a
non-pulmonary	0	1 (2.4%)
PaO2/FiO2 ratio	267.8±111.9	223.3±107.7	0.098
Shock	11 (37.9%)	24 (58.5%)	0.145
ARDS	2 (6.9%)	9 (22%)	0.108
Neutropenia	3 (10.3%)	3 (7.3%)	0.686^a
Hemodialysis	2 (6.9%)	3 (7.3%)	1^a
DIC	0	7 (17.1%)	0.036
Albumin	2.6±0.5	2.2±0.5	0.022

APACHE, Acute Physiology And Chronic Health Evaluation; SOFA, Sequential Organ Failure Assessment; PaO2, partial pressure of oxygen in arterial blood; FiO2, the fraction of inspired oxygen; ARDS, acute respiratory distress syndrome; DIC, disseminated intravascular coagulation.

Table 3.

Predictors of In-Hospital Mortality by Logistic Regression Analysis

	OR	95% CI for OR	p value
SOFA scores	1.360	1.038–1.782	0.026
Albumin	0.253	0.059–1.078	0.063
DIC	0	0-	0.999
ECOG-PS	1.655	0.855–3.204	0.135

SOFA, Sequential Organ Failure Assessment; DIC, disseminated intravascular coagulation; ECOG-PS, Eastern Cooperative Oncology Group performance status; OR, odds ratio; CI, confidence interval.

Kaplan–Meier estimates of 7, 14, and 30 day survival were 91.4%, 76.9%, and 47.5%, respectively and in-hospital survival rate was 41.4%. Patients with higher SOFA scores (≥ 7) had lower survival rates, which decreased more markedly than did those with lower SOFA scores (≤ 6) (Table 4). Kaplan–Meier analysis also revealed a significant difference between survival of patients with higher (≥ 7) versus lower (≤ 6) SOFA scores (log rank test p<0.001) (see Fig. 1).

FIG. 1.

Patient outcomes stratified by Sequential Organ Failure Assessment (SOFA) scores.

Table 4.

Patient Survival and Rate of Weaning from Ventilator, Stratified by SOFA Scores≤6 Versus≥7

		Survival rate (95% CI)
SOFA scores	n	7-day	14-day	30-day	In-hospital	Weaned from ventilator among survivors to hospital discharge
≤6	33	97.0% (80.4–100.0%)	97.0% (80.4–100.0%)	68.0% (47.3–82.0%)	63.6%	42.9%
≥7	37	83.8% (67.4–92.4%)	59.5% (42.0–73.2%)	30.3% (16.1–45.9%)	21.6%	50%
Total	70	91.4% (81.9–96.1%)	76.9% (65.1–85.1%)	47.5% (34.8–59.2%)	41.4%	44.8%

SOFA, Sequential Organ Failure Assessment.

Discussion

This study demonstrated that overall organ dysfunction as evidenced by SOFA scores is an independent mortality predictor for late-stage lung cancer patients with sepsis-related respiratory failure. Our analysis showed the patients with SOFA scores ≥7 experienced a 6.148-fold risk for mortality (95% CI: 1.545–24.465, p=0.010). As higher SOFA scores were associated with a worse prognosis, this simple index, thereby, may help inform decisions about admission to the ICU.

In a large cohort study, Taccone and colleagues reported that cancer patients were more often admitted to the ICU because of sepsis and respiratory complications than were other patients.⁶ When cancer patients face sepsis, their families and physicians often struggle to decide whether aggressive treatment should be given or not. It is not uncommon to encounter a family continuing to hope against hope that an imminently dying patient will be rescued no matter how poor that patient's condition is. However, overaggressive management may not benefit imminently dying patients, and could impose additional economic burden and cause ineffective utilization of health care resources.¹⁵ In contrast, early withdrawal of aggressive measures may prevent potentially curable patients from timely and effective treatment. How to define who may or may not benefit from intensive care has been debated for a long time; this issue involves the wishes of patients and their families, attitudes of physicians in charge, objective evaluation (e.g., cancer stage, performance status, severity of sepsis) and ethical concerns.¹⁶ Despite that admission of cancer patients to the ICU is no longer considered ineffective, it matters to find out who may benefit from such intensive care. Our study focused on advanced lung cancer patients, who were admitted to ICU for mechanical ventilator support for sepsis-related respiratory failure, a common scenario we encounter.^6,10

Compared with other studies regarding lung cancer patients with ICU admission, in-hospital mortality (58.6%) in this study was similar to that reported by Reichner et al (60%)¹⁷ and that by Soares et al (59%).¹⁸ The in-hospital mortality rate approaching 60% in the above three studies (including our results) is better than that reported several years ago by Ewer et al. (91%),³ that by Boussat et al. (75%),¹⁹ and that by Lin et al. (85.2%),²⁰ possibly implying the progress in critical care and agressiveness in treatment for cancer patients. Despite the higher percentage of sepsis among our enrollees, there was no obvious difference in hospital survival between our study and those of Reichner or Soares.^17,18 Compared with other etiology resulting in ICU admission, the probable better prognosis of sepsis in late-stage lung cancer patients may encourage us to provide more aggressive treatment for them. Nevertheless, the mortality rate was still high and the predictors for patient outcome may help us to guide further treatment.

Our data revealed that the survival rate varies with the overall severity of organ dysfunction. This is compatible with the findings reported in the other studies, regardless of measures of organ dysfunction by the number of organ failures or by SOFA/logistic organ dysfunction (LOD) scores.^7,18,21 However, enrollees in these studies were more heterogeneous, with nonseptic patients included, and there were more predictors used and combined to make a final conclusion, thereby limiting their applicability. In contrast, our study focused on late-stage lung cancer patients with sepsis-related respiratory failure, and may help management for such patients.

As higher SOFA scores were associated with poor prognosis in our study, we would suggest considering early palliative care, even hospice care, for terminal lung cancer patients with sepsis–related respiratory failure, especially for those with SOFA scores ≥7. Even though those patients survived that episode of sepsis, approximately half of them appeared to depend on a ventilator at hospital discharge, which could further impair their quality of life and prevent or delay implementation of scheduled aggressive therapy for cancer. More and more evidence has emerged to support the benefit of early palliative care in improvement of quality of life for patients with advanced cancer or for noncancer patients admitted to an ICU.^22–24 In a randomized controlled trial by Temel and colleagues, early introduction of palliative care could improve mood and quality of life, and even prolong survival for patients with metastatic non-small-cell lung cancer.²²

Limitations

There are some limitations worth noting regarding this study, including those inherent to its retrospective nature and single-center setting. Our study samples were relatively smaller in size, of pure Chinese ethnicity, and predominantly composed of male patients, which may lessen the ability of our conclusion to be generalized. As our study did not include patients with lung cancer who had been denied or whose family had foregone ICU admission, possible selection biases cannot be ruled out. This unsurveyed population may represent a substantial portion of cancer patients, which possibly differs from our enrollees in disease course or survival. However, we did not have data about this population.

In addition, sepsis in our subjects was mostly of pulmonary origin and whether our results can be applicable to patients with sepsis of other origin remains a concern. Furthermore, our study was confined to lung cancer patients. Different cancers may vary in clinical course, treatment response, and complication. Expansion of our conclusion to patients with other malignancies may warrant further investigation.

Conclusions

In conclusion, our study showed that the SOFA score is an independent predictor of outcome for late-stage lung cancer patients with sepsis-related respiratory failure. All patients with advanced lung cancer should receive palliative care, and those with sepsis-related respiratory failure and high SOFA scores are suggested for hospice care.

Footnotes

Acknowledgments

We thank Dr. Chen Yung-Tai, a physician of Taipei City Hospital, Heping-Fuyou branch, for his assistance with statistics. The Institutional Review Board of Taipei Veterans General Hospital approved the study (VGHIRB No. 201012036IC).

Author Contributions

All authors had access to the data and played a role in writing this article.

Author Disclosure Statement

No competing financial interests exist.

References

Ferlay

, Shin

, Bray

, Forman

, Mathers

, Parkin

. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer, 2010; 127:2893–2917.

Schiller

, Harrington

, Belani

, Langer

, Sandler

, Krook

, Zhu

, Johnson

. Eastern Cooperative Oncology Group. Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. N Engl J Med, 2002; 346:92–98.

Ewer

, Ali

, Atta

, Morice

, Balakrishnan

. Outcome of lung cancer patients requiring mechanical ventilation for pulmonary failure. JAMA, 1986; 256:3364–3366.

Mendoza

, Lee

, Marik

. The hospital-survival and prognostic factors of patients with solid tumors admitted to an ICU. Am J Hosp Palliat Care, 2008; 25:240–243.

Boussat

, El'rini

, Dubiez

, Depierre

, Barale

, Capellier

. Predictive factors of death in primary lung cancer patients on admission to the intensive care unit. Intensive Care Med, 2000; 26:1811–1816.

Taccone

, Artigas

, Sprung

, Moreno

, Sakr

, Vincent

. Characteristics and outcomes of cancer patients in European ICUs. Crit Care, 2009; 13:R15.

Toffart

, Minet

, Raynard

, Schwebel

, Hamidfar–Roy

, Diab

, Quetant

, Moro–Sibilot

, Azoulay

, Timsit

. Use of intensive care in patients with nonresectable lung cancer. Chest, 2011; 139:101–108.

Darmon

, Azoulay

. Critical care management of cancer patients: cause for optimism and need for objectivity. Curr Opin Oncol, 2009; 21:318–326.

Sharma

, Freeman

, Zhang

, Goodwin

. Trends in end-of-life ICU use among older adults with advanced lung cancer. Chest, 2008; 133:72–78.

10.

Pastores

, Voigt

. Acute respiratory failure in the patient with cancer: diagnostic and management strategies. Crit Care Clin, 2010; 26:21–40.

11.

Roques

, Parrot

, Lavole

, Ancel

, Gounant

, Djibre

, Fartoukh

. Six-month prognosis of patients with lung cancer admitted to the intensive care unit. Intensive Care Med, 2009; 35:2044–2050.

12.

Oken

, Creech

, Tormey

, Horton

, Davis

, McFadden

, Carbone

. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol, 1982; 5:649–655.

13.

Vincent

, Moreno

, Takala

, Willatts

, De Mendonça

, Bruining

, Reinhart

, Suter

, Thijs

. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med, 1996; 22:707–710.

14.

Working group on “sepsis-related problems” of the European Society of Intensive Care Medicine. Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units: results of a multicenter, prospective study. Crit Care Med, 1998; 26:1793–1800.

15.

Kutikova

, Bowman

, Chang

, Long

, Obasaju

, Crown

. The economic burden of lung cancer and the associated costs of treatment failure in the United States. Lung Cancer, 2005; 50:143–154

16.

Soubani

, Ruckdeschel

. The outcome of medical intensive care for lung cancer patients: the case for optimism. J Thorac Oncol, 2011; 6:633–638.

17.

Reichner

, Thompson

, O'Brien

, Kuru

, Anderson

. Outcome and code status of lung cancer patients admitted to the medical ICU. Chest, 2006; 130:719–723.

18.

Soares

, Darmon

, Salluh

, Ferreira

, Thiéry

, Schlemmer

, Spector

, Azoulay

. Prognosis of lung cancer patients with life-threatening complications. Chest, 2007; 131:840–846.

19.

Boussat

, El'rini

, Dubiez

, Depierre

, Barale

, Capellier

. Predictive factors of death in primary lung cancer patients on admission to the intensive care unit. Intensive Care Med, 2000; 26:1811–1816.

20.

Lin

, Tsai

, Huang

, Hsu

, Wang

, Tsao

, Lin

. Outcome of lung cancer patients with acute respiratory failure requiring mechanical ventilation. Respir Med, 2004; 98:43–51.

21.

Adam

, Soubani

. Outcome and prognostic factors of lung cancer patients admitted to the medical intensive care unit. Eur Respir J, 2008; 31:47–53.

22.

Temel

, Greer

, Muzikansky

, Gallagher

, Admane

, Jackson

, Dahlin

, Blinderman

, Jacobsen

, Pirl

, Billings

, Lynch

. Early palliative care for patients with metastatic non-small-cell lung cancer. N Engl J Med, 2010; 363:733–742.

23.

Greer

, Pirl

, Jackson

, Muzikansky

, Lennes

, Heist

, Gallagher

, Temel

. Effect of early palliative care on chemotherapy use and end-of-life care in patients with metastatic non-small-cell lung cancer. J Clin Oncol, 2012; 30:394–400.

24.

O'Mahony

, McHenry

, Blank

, Snow

, Eti Karakas

, Santoro

, Selwyn

, Kvetan

. Preliminary report of the integration of a palliative care team into an intensive care unit. Palliat Med, 2010; 24:154–165.