| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Medicine (Dr. Skaug), Haugesund Hospital, Health Region of Fonna, Haugesund; Centre for Clinical Research (Mr. Eide), Haukeland University Hospital, and Section for Epidemiology and Medical Statistics, Department of Public Health and Primary Health Care, University of Bergen, Bergen; and Department of Thoracic Medicine (Dr. Gulsvik), Institute of Medicine, University of Bergen, Bergen, Norway.
Correspondence to: Knut Skaug, MD, Haugesund Hospital, Health Region of Fonna, PO Box 2170, N-5104 Haugesund, Norway; e-mail: dr.knut{at}skaug.no
Abstract
Background: There is little knowledge concerning the prevalence and predictors of symptoms in the terminal stage of lung cancer.
Methods: We examined, retrospectively, all cases of lung cancer diagnosed from 1990 to 1996 in a defined hospital area in Norway. All medical records from general practitioners, nursing homes, and hospitals were investigated. A total of 271 cases were diagnosed, and 247 of 253 deaths (98%) were analyzed.
Results: In the terminal 8 weeks, pain was recorded in 85% of the patients, psychological symptoms (anxiety, insomnia, and/or depression) in 71%, dyspnea in 54%, neurologic symptoms in 28%, cough in 24%, nausea in 21%, and hemoptysis in 9%. Young age (p = 0.02) and small cell lung carcinoma (p = 0.03) were risk factors for psychological symptoms. Terminal dyspnea was more frequent in patients with stage III (p = 0.002) and nausea in stage IV (p = 0.02) at the time of diagnosis, while cough (p = 0.04) occurred more often in non-small cell lung carcinoma. Terminal pain was independent of gender, age, performance status, stage, and histology.
Conclusion: In a community health service encompassing all lung cancer patients, pain, psychological symptoms, and dyspnea were frequent complaints in the terminal phase. Terminal dyspnea and nausea were associated with staging at the time of diagnosis, and terminal cough and nausea were associated with histology.
Key Words: critical care • epidemiology, pulmonary • lung cancer
The mortality rate of lung cancer is high, and there has been little improvement in the long-term survival over the last 40 years.123 In addition, information is lacking whether sex, age, initial stage and performance status, and histology may predict the presence of serious symptoms in the terminal stage. In previous community studies of patients dying from lung cancer, 78% displayed dyspnea and 85% had pain in the final year of life,4 while depression was observed in 33% of patients with lung cancer referred to palliative treatments centers.5 In the first study,4 the participation rate was only 69%, while the latter5 was comprised of patients from three lung cancer trials with very different inclusion criteria. Comparison of these studies is difficult due to these different inclusion criteria and also to information gathered at various points during the course of the disease. Quality of life measurements give incomplete information at the end stage because many patients are unable to fill out the questionnaires.6 Therefore, information is not available concerning the incidence and prevalence of major symptoms in the terminal care of lung cancer patients from a population-based study.7
In this retrospective study, we collected information concerning the frequency of symptoms at the end stage of all lung cancer patients within a geographic area of Norway. Furthermore, we have examined if these symptoms could have been predicted at the time of diagnosis by gender, age, extension of the neoplasm, functional performance, and histology of the tumor.
Materials and Methods
Study Location
Haugalandet is located in southwest Norway and includes the city of Haugesund and nine surrounding municipalities. In 1996, the hospital district consisted of 100,724 inhabitants, of whom 77,337 were > 15 years of age.
Cases
The diagnosis of lung cancer was confirmed by histology or convincing radiologic signs (chest radiography, CT). Included in this survey were the following: (1) all new patients in the Norwegian Cancer Registry8 with lung cancer in International Classification of Diseases, Seventh Revision (1990 to 1992) and International Classification of Diseases, Ninth Revision (1993 to 1996) living in Haugalandet; and (2) all new lung cancer patients admitted during 1990 to 1996 to Haugesund Hospital. Exclusion criteria were as follows: (1) patients with lung cancer in the same period but not residents of Haugalandet; (2) patients alive as of December 1, 2003; and (3) patients with no records of the terminal stage of the disease. The study was approved by the Regional Committee for Medical Research Ethics of West Norway.
Case Record Form
The case record forms included data at the time of diagnosis on sex; age; description of primary tumor (T); regional lymph nodes (N); distant metastasis (M); stage group according to the International System for Staging Lung Cancer9; and histology and initial performance status scale according to the World Health Organization (WHO).10 Signs, symptoms at the time of diagnosis and in the terminal 8 weeks, treatment, place of death (home, hospitals, and nursing homes), and concurrent diseases were recorded.
Terminal Symptoms
The terminal stage of lung cancer was defined as the last 8 weeks of life, and relevant terminal symptoms were extracted from patient records. Pain was graded according to its intensity: if the symptoms required peripheral analgesics, opioids, or a morphine infuser. The presence and possible causes of dyspnea (central stenosis when a tumor was observed on bronchoscopy, pleural fluid observed on chest radiographs, or other reasons) was recorded. Psychological symptoms were recorded in patients files and in addition prescribed drugs according to the Anatomical Therapeutic Chemical Classification System,11 in which drugs are localized into different groups according to the organ or system on which they act: depression, when administered antidepressive medication (N 06); anxiousness, when administered diazepam (N05 B A 01) or other anxiolytic treatments (N05 B A 04); or insomnia, when administered hypnotics (N 05 C). Dizziness, headache, or signs of paresis were defined as neurologic symptoms. Additional terminal symptoms recorded were nausea, cough, and hemoptysis. The number of weeks in the terminal 8 weeks treated for pain, psychological symptoms, and dyspnea was recorded. The case record form was tested in a pilot study of 20 patients by two consultants in internal medicine. The wording of the questions was thereafter improved, a semiquantification of symptom load was added, and imprecise questions were deleted. Dates of death were collected from the Cancer Registry of Norway.8 The records from hospitals, nursing homes, and general practitioners were reviewed randomly by three physicians following written instructions provided by the principal author.
Statistical Methods
The case record forms were scanned into a database in Microsoft Excel Worksheet (Microsoft; Redmond, WA) and then imported into a statistical program (SPSS 10.5; SPSS; Chicago, IL). Gender, age, initial stage and performance status, and histology were analyzed as predictors for symptoms such as pain, dyspnea, nausea, cough, psychological symptoms, and neurologic signs, respectively, using two-way contingency tables. Finally, backwards multiple logistic regression analysis was performed for each symptom with respect to the potential predictors and adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were calculated.
Results
Demography
There were 271 incident lung cancer patients from 1990 to 1997. Eighteen patients (6.6%) were alive by December 1, 2003. The diagnosis was confirmed by histology in 218 patients (89%). Information on symptoms and treatment in the terminal 8 weeks were collected from 247 of the 253 patients (98%) who died within this period of surveillance. One patient died abroad, and the records of five patients were not available. Only 53 of the 247 deaths (22%) were in women. The distribution of age, stage, performance status, and histology at the time of initial diagnosis (Table 1
) did not differ between genders (p > 0.05). Of 56 patients with small cell lung carcinoma (SCLC) 27 patients had localized disease corresponding to stage I-III, and 29 patients had extensive disease with metastasis corresponding to stage IV. The average age was 68.5 years (SD, 10.1 years; range, 38 to 89 years). Initial antineoplastic treatment (surgery, radiotherapy, and/or chemotherapy) was administered to 60% of the patients. Lung cancer was the underlying cause of death in 227 patients and was contributory to death in 15 of the remaining 20 patients. The average time from beginning of symptoms to diagnosis was 89 days.
|
Diagnosis for 29 patients (11%) was based on findings from chest radiographs and CT scans. The 29 lung cancer patients who did not have histologic confirmation were older (mean age, 74.3 years; p = 0.005), and their performance status was poorer (p < 0.001) compared to the 218 patients with confirmed histology. The median survival time of these 29 patients was only 74 days, compared to 123 days for an age-matched group with histologic confirmation.
Terminal Symptoms
Univariate Analysis:
The prevalence of symptoms by age in the terminal 8 weeks is shown in Table 2
. There was no difference between genders in relation to any of these symptoms. Pain was present in 211 patients (85%). Dyspnea occurred in 134 patients (54%), and the prevalence varied between lung cancer stages (p = 0.002), being most frequent if the patient was in stage III (Fig 1
). Twenty-eight percent of these patients had central stenosis visible at bronchoscopy, and 33% had pleural fluid. Nausea was present in a higher proportion of patients with SCLC (41%), as opposed to patients with non-small cell lung carcinoma (NSCLC) [25%; p < 0.01]. This varied with stage, being most frequent in stage IV with disseminated disease (Fig 1). Forty-six of 56 patients with SCLC received chemotherapy. Twenty of these patients had terminal nausea. Only 12 of 162 patients with NSCLC received chemotherapy, and 4 patients had terminal nausea. Cough was more frequent in patients with NSCLC (36%; Fig 1) than in patients with SCLC (28%; p = 0.04). Hospitalized patients had cough more frequently (33%) than nonhospitalized patients (19%; p = 0.01).
|
|
The duration of symptomatic treatment varied among symptoms (Fig 2 ). The majority of the patients with pain, dyspnea, and neurologic symptoms had symptoms all 8 weeks. Patients with psychological symptoms had equal numbers with symptoms for 8 weeks as those with a shorter time span.
|
|
In the terminal stage of lung cancer, 99% of the patients had at least one or more symptoms of pain, dyspnea, nausea, cough, hemoptysis, and psychological and neurologic difficulties. The most frequent symptoms were pain, psychological manifestations, and/or dyspnea. One fourth of patients had nausea, cough, and/or neurologic symptoms, and only one tenth displayed hemoptysis. Two thirds of the terminal lung cancer patients had three or more symptoms.
The strength of this study is that 98% of all lung cancer patients in a defined hospital area were included and 89% had a histologic confirmation, which is higher than previous studies.3 The cohort of patients was followed up from 5 to 11 years, and we were able to obtain information concerning the terminal stage of lung cancer in 98% of those who died.
Comparison between previous studies shows inconsistencies that are reflected in varying results (Table 4 ). First, selection criteria for lung cancer patients and the representation of the cases are different.45712 Second, the methods of assessing symptoms vary considerably.41213 Third, we question how representative and accurate the self-reported answers are compared to the physician’s report at the end stage. Self-reported questionnaires are dependent on full cooperation and alertness of the patient and thus are not available for a high proportion of the patients in terminal care.141516 As the disease progresses, there is a high dropout rate from the quality of life evaluations. Fourth, different time windows in the assessment of symptoms during the course of lung cancer also make it difficult to compare studies.
|
In our study, 29% of the patients received antidepressive medication. In agreement with this, Hopwood and Stephens5 observed depression in 33% of lung cancer patients using the Hospital Anxiety and Depression Scale. Furthermore, they observed a trend for higher rates of depression in the younger age groups, which is in agreement with our findings in which age < 65 years predicted more psychological symptoms. However, in contrast to our study they found that depression was strongly predicted by poor initial performance status.5
Patients with SCLS often have paraneoplastic syndromes with production of peptide hormones,17 which can be a cause of nausea.18 This may be an explanation why SCLC was a predictor for terminal nausea. An additional cause may be the association to chemotherapy administered early in the course of the disease to the majority of these patients. It is known that patients who have nausea during chemotherapy react to later stimuli that remind them of their previous experience.19 Stage IV disease was a predictor for terminal nausea, which may also be explained by metastasis to both the liver and the brain.18
A retrospective study recording terminal symptoms has limitations with regard to recall and completeness of symptom records. However, according to Norwegian legislation, symptoms demanding treatment have to be noted in the patient’s records. This applies to hospitals, nursing homes, and a physician’s office practice.20 This increases our confidence that all serious symptoms in our study have been recorded, but less serious complaints may have been omitted. The information concerning anorexia and weight loss was incomplete and therefore not included in our study.
None of the possible predicting factors examined in our study (Table 3) were associated with a higher use of analgesics in the terminal care. This is in contrast to another cancer study21 in which there is an inverse relationship between the intensity of pain treatment and age. The reason that the hospitalized patients in our study were administered more opioids is not clear. There were no differences in the demographic data, histology, or stage of tumor between patients with terminal care within or outside the hospital. However, the hospitalized patients may have had more pain, or the staff in the hospitals may have been more confident and experienced concerning the use of opioids than external health-care workers.
In conclusion, those who provide health care have to be aware of the high frequency of pain, dyspnea, and psychological symptoms in the terminal stage of lung cancer. The high prevalence of symptoms toward the end of life requires frequent follow-up. Dyspnea and nausea are more often observed if there is extensive disease at the time of diagnosis. Nausea is present if the diagnosis is SCLC, and psychological symptoms are increased if the patient is < 65 years old at the time of diagnosis. General practitioners and consultants should be aware of predictors for these symptoms and should be trained in optimal palliative treatment for the symptoms.22
Acknowledgements
We thank Sverre Fluge, MD, and Sveinung Sørhaug, MD, for participating in reviewing the patient records; Randi Eikeland, Data Manager, Centre for Clinical Cancer Research, Haukeland University Hospital, for scanning the data from the case record forms into a database; and Doris Gundersen, PhD, Haugesund Hospital, for reviewing the manuscript.
Footnotes
Abbreviations: CI = confidence interval; NSCLC = non-small cell lung carcinoma; OR = odds ratio; SCLC = small cell lung carcinoma; WHO = World Health Organization
This work was performed at Haugesund Hospital.
The authors have no conflicts of interest to disclose.
Received for publication May 12, 2006. Accepted for publication August 10, 2006.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
