Advertisement

High Lung Cancer Incidence in Heavy Smokers Following Hospitalization due to Pneumonia

Published:November 06, 2015DOI:https://doi.org/10.1016/j.amjmed.2015.10.030

      Abstract

      Introduction

      The rate of lung cancer incidence following pneumonia in heavy smokers is unknown. Heavy smokers hospitalized due to community-acquired pneumonia might be at high risk for subsequent lung cancer. The primary objective of this study was to determine lung cancer incidence in this high-risk population.

      Patients and Methods

      This was a single-center, retrospective cohort study that included heavy smokers hospitalized due to community-acquired pneumonia between January 1, 2007 and December 31, 2011 in Beilinson hospital, a large community hospital and tertiary center. Patients were identified by International Classification of Diseases, Ninth Revision coding from the hospital's registry. Two physicians reviewed every patient's medical file for patient demographics, smoking history, lung cancer risk factors, and anatomical location of pneumonia. Data were cross-checked with the database at the national cancer registry for new diagnoses of cancer.

      Results

      There were 381 admissions for community-acquired pneumonia included in the final analysis. Thirty-one cases (8.14%; 95% confidence interval [CI], 5.9%-11.2%) of lung cancer were diagnosed during the first year after hospitalization. Lung cancer incidence was significantly higher in patients who had upper-lobe pneumonia (23.8%; 95% CI, 14.9%-40%). Lung cancer was located within the lobe involved by the pneumonia in 75.8% of patients.

      Conclusions

      A high lung cancer rate was found in heavy smokers admitted due to community-acquired pneumonia. The association was especially strong for patients with upper-lobe pneumonia. Screening with chest computed tomography should be strongly considered for these patients.

      Keywords

      Clinical Significance
      • One-year cumulative incidence of lung cancer following hospitalization due to pneumonia in heavy smokers was 8.14%.
      • Patients with upper-lobe infiltrates had a significantly higher probability of subsequent lung cancer. Lung cancer was located within a lobe involved by pneumonia in 75.8% of patients.
      • We believe we have identified a patient population that might benefit from early screening with chest computed tomography scan.
      Lung cancer is the leading cause of cancer mortality in the US, associated with a 5-year survival of 17%.
      • Humphery L.L.
      • Deffebach M.
      • Pappas M.
      • et al.
      Screening for lung cancer with low dose computed tomography: a systemic review to update the US preventive task force recommendations.
      The most important risk factor for lung cancer is smoking, which causes approximately 85% of all lung cancer cases.1 Only 15% of patients are diagnosed at an early stage.
      • Humphery L.L.
      • Deffebach M.
      • Pappas M.
      • et al.
      Screening for lung cancer with low dose computed tomography: a systemic review to update the US preventive task force recommendations.
      The National Lung Screening Trial (NLST) demonstrated that screening heavy smokers with low-dose computed tomography reduces mortality from lung cancer.
      • Aberle D.R.
      • Adams A.M.
      • Berg C.D.
      • et al.
      The National Lung Screening Trial Research Team
      Reduced lung-cancer mortality with low-dose computed tomographic screening.
      Following the NLST, the United States Preventive Services Task Force and the Centers for Medicare and Medicaid Services recommend annual screening for lung cancer with low-dose computed tomography in adults who have a 30 pack-year smoking history and currently smoke or have quit within the past 15 years, ages 55-80 (US Preventive Services Task Force) or 55-77 years (Centers for Medicare and Medicaid Services).

      U.S. Preventive Services Task Force (USPSTF). Final recommendation statement. Lung cancer: screening. Available at: http://www.uspreventiveservicestaskforce.org/Page/Document/RecommendationStatementFinal/lung-cancer-screening. Accessed September 2, 2015.

      Centers for Medicare and Medicaid Services
      Decision memo for screening for lung cancer with low dose computed tomography.
      However, there are numerous controversies about this broad recommendation, regarding both medical and cost-effective issues.
      • Bach P.B.
      • Mirkin J.N.
      • Oliver T.K.
      • et al.
      Benefits and harms of CT screening for lung cancer: a systematic review.
      • Field J.K.
      • Oudkerk M.
      • Pedersen J.H.
      • Duffy S.W.
      Prospects for population screening and diagnosis of lung cancer.
      • Woolf S.H.
      • Harris R.P.
      • Campos-Outcalt D.
      Low-dose computed tomography screening for lung cancer: how strong is the evidence?.
      • Patz Jr., E.F.
      • Pinsky P.
      • Gatsonis C.
      • et al.
      NLST Overdiagnosis Manuscript Writing Team
      Overdiagnosis in low-dose computed tomography screening for lung cancer.
      • Black W.C.
      • Gareen I.F.
      • Soneji S.S.
      • et al.
      National Lung Screening Trial Research Team
      Cost-effectiveness of CT screening in the National Lung Screening Trial.
      There is a need to improve the risk assessment of individuals in order to screen only those who are truly at high risk for this malignancy.
      Pneumonia is a common pulmonary infection, accounting for more than 1.1 million annual hospitalizations in the US.

      Centers for Disease Control and Prevention (CDC). National Center for Health Statistics. Pneumonia. Available at: http://www.cdc.gov/nchs/fastats/pneumonia.htm. Accessed September 2, 2015.

      Several professional societies recommend a follow-up chest radiograph 4-8 weeks after treatment for pneumonia, to exclude an underlying lung cancer.
      • Mandell L.A.
      • Marrie T.J.
      • Grossman R.F.
      • Chow A.W.
      • Hyland R.H.
      Canadian guidelines for the initial management of community-acquired pneumonia: an evidence-based update by the Canadian Infectious Diseases Society and the Canadian Thoracic Society. The Canadian Community-Acquired Pneumonia Working Group.
      • Niederman M.S.
      • Mandell L.A.
      • Anzueto A.
      • et al.
      American Thoracic Society
      Guidelines for the management of adults with community-acquired pneumonia. Diagnosis, assessment of severity, antimicrobial therapy, and prevention.
      • Ramsdell J.
      • Narsavage G.L.
      • Fink J.B.
      American College of Chest Physicians' Home Care Network Working Group
      Management of community-acquired pneumonia in the home: an American College of Chest Physicians clinical position statement.
      • Lim W.S.
      • Baudouin S.V.
      • George R.C.
      • et al.
      Pneumonia Guidelines Committee of the BTS Standards of Care Committee
      BTS guidelines for the management of community acquired pneumonia in adults: update 2009.
      Published data supporting this practice are scarce, with several studies documenting lung cancer incidence of approximately 1% at 90 days and 2%-3% at 5 years following pneumonia.
      • Tang K.L.
      • Eurich D.T.
      • Minhas-Sandhu J.K.
      • Marrie T.J.
      • Majumdar S.R.
      Incidence, correlates, and chest radiographic yield of new lung cancer diagnosis in 3398 patients with pneumonia.
      • Marrie T.J.
      Pneumonia and carcinoma of the lung.
      • Holmberg H.
      • Kragsbjerg P.
      Association of pneumonia and lung cancer: the value of convalescent chest radiography and follow-up.
      • Søyseth V.
      • Benth J.S.
      • Stavem K.
      The association between hospitalisation for pneumonia and the diagnosis of lung cancer.
      However, Mortensen et al
      • Mortensen E.M.
      • Copeland L.A.
      • Pugh M.J.
      • et al.
      Diagnosis of pulmonary malignancy after hospitalization for pneumonia.
      found a post-pneumonia lung cancer rate of 9.2% at 5 years, with 27% of them (representing 2.48%) diagnosed within 90 days.
      • Mortensen E.M.
      • Copeland L.A.
      • Pugh M.J.
      • et al.
      Diagnosis of pulmonary malignancy after hospitalization for pneumonia.
      The reason for this discrepancy is probably the different patient populations studied; while most studies included many young patients and few smokers, Mortensen et al
      • Mortensen E.M.
      • Copeland L.A.
      • Pugh M.J.
      • et al.
      Diagnosis of pulmonary malignancy after hospitalization for pneumonia.
      studied a significantly older patient population (mean age of 77 years) of whom 36% were current smokers.
      Pneumonia is considerably more common in smokers as compared with nonsmokers.
      U.S. Department of Health and Human ServicesPublic Health ServiceOffice of the Surgeon General
      The Health Consequences of Smoking—50 Years of Progress. A Report of the Surgeon General 2014.
      • Almirall J.
      • Bolibar I.
      • Serra-Prat M.
      • et al.
      New evidence of risk factors for community-acquired pneumonia: a population-based study.
      • Nuorti J.P.
      • Butler J.C.
      • Farley M.M.
      • et al.
      Cigarette smoking and invasive and invasive pneumococcal disease. Active Bacterial Core Surveillance Team.
      It is also a common complication of lung cancer.
      • Nichols L.
      • Saunders R.
      • Knollmann F.D.
      Causes of death of patients with lung cancer.
      Thus, diagnosis of pneumonia is a potential marker for an occult lung cancer in smokers. To the best of our knowledge, no study has evaluated the incidence of post-pneumonia lung cancer in a homogenous population of patients at high risk for this malignancy. We hypothesized that heavy smokers over 40 years of age hospitalized due to community-acquired pneumonia may represent a patient population at high risk for lung cancer. To test our hypothesis we conducted a single-center retrospective cohort study to determine the incidence of lung cancer during 1 year following hospitalization due to community-acquired pneumonia in this patient population.

      Methods

      The study was approved by the hospital's Institutional Review Board. The study population included all patients hospitalized due to community-acquired pneumonia between January 1, 2007 and December 31, 2011 in Beilinson Hospital, Rabin Medical Center, Israel (a large community hospital and a tertiary center). Only patients older than 40 years of age with a smoking history of at least 30 pack-years were included. Former smokers were included if they had quit within the previous 15 years. The patients were identified by International Classification of Diseases, Ninth Revision coding from the hospital's registry using codes 480-486 for pneumonia, 305.1 for current smokers, and V15.82 for former smokers. Two physicians reviewed patients' medical files to ascertain that the diagnosis of pneumonia has been made according to the National Center for Health Statistics definitions.

      Centers for Disease Control and Prevention (CDC). CDC/NHSN surveillance definitions for specific types of infection. Available at: http://www.cdc.gov/nhsn/PDFs/pscManual/17pscNosInfDef_current.pdf. Accessed September 2, 2015.

      Patients with hospital-acquired pneumonia, ventilator-associated pneumonia, witnessed aspiration pneumonia, death during hospitalization, prior diagnosis of lung cancer, lung metastasis, or pulmonary transplantation were excluded from our study.
      Collected data included demographics, smoking history, diagnosis of chronic obstructive pulmonary disease, alcohol and drug abuse, and family and personal history of cancer. Pneumonia in the previous year was classified as recurrent pneumonia. Two admissions due to pneumonia separated by a year or more were regarded as 2 discrete events. The anatomical location of the pneumonia was noted. The study cohort was linked by national ID number with the Israel National Cancer Registry (INCR) database in order to identify new cancer diagnoses. The INCR covers the entire Israeli population (approximately 8 million), and at the time this study was conducted was complete through December 31, 2011. Lag time from diagnosis of pneumonia to cancer diagnosis was calculated. Stage at diagnosis was noted from patient files, when available. Data regarding patients' vital status until October 1, 2014 were collected from health maintenance organizations.

      Statistical Analysis

      Chi-squared or Fisher's exact test were used to compare categorical variables between study groups. Student's t test was used to compare continuous variables between study groups.
      Lung cancer cumulative incidence and confidence intervals (CI) were assessed via the Cox proportional hazards model with the Fine and Gray methodology
      • Fine J.P.
      • Gray R.J.
      A proportional hazards model for the subdistribution of competing risk.
      of dealing with cancer-free death as a competing risk.

      Results

      There were 787 admissions due to pneumonia for patients who were smokers between January 1, 2007 and December 31, 2011. There were 406 events excluded for various reasons (Figure). Final analysis included 381 admissions of 363 different patients (18 patients had 2 admissions for pneumonia more than 1 year apart). Patient characteristics are detailed in Table 1. Median follow-up was 24 months for new malignancies (INCR data) and 46 months regarding survival (health maintenance organization's data). Fifty-three percent of the cohort (202 patients) died during follow-up. There were 53 (13.9%) new cases of cancer during the follow-up period: 33 (8.66%) lung cancers and 20 (5.25%) cases of other malignancies. The most common malignancy excluding lung cancer was head and neck squamous cell carcinoma, accounting for 5 (25%) of the new non-lung cancers. In the year following diagnosis of pneumonia, the cumulative incidence of lung cancer was 8.14% (95% CI, 5.9%-11.2%).
      Figure thumbnail gr1
      FigureCauses of patient exclusion. CDC = Centers for Disease Control and Prevention; NHSN = National Healthcare Safety Network.
      Table 1Lung Cancer Risk According to Demographic and Clinical Data
      Study PopulationPatients Diagnosed with Lung CancerLung Cancer Incidence (95% CI)P Value
      No. of patients381338.7% (6.2%-11.9%)
      Sex1
       Male283 (74.3%)25 (75.8%)8.8% (6%-12.7%)
       Female98 (25.7%)8 (24.2%)8.2% (4.2%-15.3%)
      Mean age (range)66 (40-94, SD 12.1)66.1 (43-83, SD 12.4).95
      Mean pack years (range)60.9 (30-275, SD 35.1)64.8 (30-150, SD 64.8).49
      Active smokers266 (69.8%)25 (75.8%)9.4% (6.5%-13.5%).55
      Past smokers115 (30.1%)8 (24.2%)7% (3.6%-13.1%)
      Patients with documented pneumonia in previous year25 (6.6%)4 (12.1%)16% (6.4%-34.7%).27
      No pneumonia previous year356 (93.4%)29 (87.9%)8.2% (5.7%-11.5%)
      COPD199 (52.2%)20 (60.6%)10% (6.6%-15%).36
      No COPD182 (47.8%)13 (39.4%)7.2% (4.2%-11.8%)
      Alcohol abuse31 (8.1%)2 (6.1%)6.5% (1.8%-20.7%)1
      No alcohol abuse350 (91.9%)31 (93.9%)8.9% (6.3%-12.3%)
      Drug abuse8 (2.1%)0 (0%)0% (0%-32.4%)1
      No drug abuse373 (97.9%)33 (100%)8.9% (6.4%-12.2%)
      Personal history of cancer58 (15.2%)2 (6.1%)3.5% (1%-11.7%).2
      No personal history of cancer323 (84.8%)31 (93.9%)9.6% (6.9%-13.3%)
      Family history of cancer30/102 (29.4%)
      Data were missing regarding family history of cancer (279 patients).
      7/30 (23.3%)23.3% (11.8%-40.1%).037
      No family history of cancer72/102 (70.6%)23/30 (76.7%)32% (22.3%-43.4%)
      Family history of lung cancer6 (1.6%)2 (6.1%)33.3% (9.7%-70%).09
      No family history of lung cancer375 (98.4%)31 (93.9%)8.3% (5.9%-11.5%)
      CI = confidence interval; COPD = chronic obstructive pulmonary disease; SD = standard deviation.
      Data were missing regarding family history of cancer (279 patients).
      Patients who were diagnosed with lung cancer did not differ significantly with respect to demographic factors or medical history from those who did not develop lung cancer (Table 1). However, these patients were significantly more likely to have a family history of cancer.
      New lung cancer diagnoses were temporally related to hospitalization due to community-acquired pneumonia in our cohort. Thirty-one (93.9%) new lung cancers and seven (35%) non-lung cancers were diagnosed during the first year following hospitalization (Table 2, P = .0001). Mean lag time between pneumonia and diagnosis of lung cancer and non-lung cancer was 101 days (SD 145.2) and 554.9 days (SD 449.3), respectively (P = .0003).
      Table 2No. of Patients with Lung Cancer and Non-lung Cancer Diagnoses According to Lag Time (Days) From Pneumonia
      Days from Pneumonia to Cancer DiagnosisNo. of Patients with Post-pneumonia CancerLung CancerNon-lung Cancer
      Total533320
      1-3010 (18.9%)9 (27.3%)1 (5%)
      31-9018 (34%)15 (45.6%)3 (15%)
      91-36510 (18.9%)7 (21.2%)3 (15%)
      365+15 (28.3%)2 (6.1%)13 (65%)
      There was a significant association between pneumonia location on chest radiograph and risk of subsequent lung cancer. Patients with upper-lobe infiltrates had a significantly higher probability of subsequent lung cancer (Table 3). Lung cancer was located within a lobe involved by pneumonia in 25 of 33 (75.8%) of patients. No patient with bilateral pneumonia was diagnosed with subsequent lung cancer.
      Table 3Pneumonia Location and Subsequent 1-Year Risk of Lung Cancer
      Pneumonia LocationNo. of PatientsNo. of Patients Subsequently Diagnosed with Lung CancerNo. of Patients Not Diagnosed with Lung Cancer% of Patients Subsequently Diagnosed with Lung Cancer (95% CI)P Value
      P values are for subsequent lung cancer incidence according to pneumonia locations, as compared with bilateral pneumonia.
      Total381313508.14% (5.9%-11.2%)
      RUL2381532.6% (18.5%-57.3%)<.001
      RML464428.2% (2.9%-22.8%).12
      RLL11251074.2% (2%-8.8%).33
      LUL2442015.4% (6.3%-37.7%).02
      LLL10961036.8% (3.6%-12.8%).34
      Upper-lobe pneumonia (RUL+LUL)47123523.8% (14.9%-40%)<.001
      Lower lobe pneumonia (LLL+RLL)221112105.5% (3.3%-9.2%).37
      Pneumonia involving several lobes of one lung3042612.5% (5.3%-29.6%).036
      Bilateral pneumonia370370% (0%-9.4%)
      CI = confidence interval; LLL = left lower lobe; LUL = left upper lobe; RLL = right lower lobe; RML = right middle lobe; RUL = right upper lobe.
      P values are for subsequent lung cancer incidence according to pneumonia locations, as compared with bilateral pneumonia.
      Data regarding histology and staging for patients diagnosed with lung cancer are provided in Table 4. Patients with different histological subtypes of lung cancer had similar demographic, clinical, radiological, and prognostic data (data not shown). Patients diagnosed with lung cancer shortly (1-30 days) following pneumonia tended to have metastatic disease more often than patients with a longer lag time. Among these patients, 8 of 9 patients (89%) had metastatic disease, compared with 11 of 20 (55%) for patients with a 31-365 days lag; however, this finding did not reach statistical significance (P = .11).
      Table 4Stage at Diagnosis According to Histology
      HistologyNo. of Patients1A3A4LDEDUnknown
      NSCLC NOS11 (100%)
      Adenocarcinoma125 (41.7%)6 (50%)1 (8.3%)
      SCC71 (14.3%)1 (14.3%)4 (57.1%)1 (14.3%)
      SCLC122 (16.7%)10 (83.3%)
      Neuroendocrine lung cancer11 (100%)
      ED = extensive disease; LD = limited disease; NOS = not otherwise specified; NSCLC = non-small-cell lung cancer; SCC = squamous cell carcinoma; SCLC = small cell lung cancer.
      Thirty of the 33 patients diagnosed with lung cancer died during follow-up. For patients diagnosed with lung cancer within 365 days of pneumonia, there was no correlation between length of time to cancer diagnosis and overall survival (P = 1) or median survival (P = .63). Median survival for patients diagnosed with lung cancer within 365 days of pneumonia was approximately 9 months. Median survival for patients not diagnosed with lung cancer was 52 months.

      Discussion

      Lung cancer incidence during 1 year following hospitalization for pneumonia in heavy smokers was 8.14% (95% CI, 5.9%-11.2%). To the best of our knowledge, this is the highest reported rate of new lung cancer in the year following pneumonia. We believe our results differ from the majority of earlier reports due to the high-risk population analyzed in our study, representing the well-documented lung cancer risk among heavy smokers, and the susceptibility of heavy smokers with lung cancer to pneumonia. Furthermore, follow-up duration was longer than in most previous studies.
      • Tang K.L.
      • Eurich D.T.
      • Minhas-Sandhu J.K.
      • Marrie T.J.
      • Majumdar S.R.
      Incidence, correlates, and chest radiographic yield of new lung cancer diagnosis in 3398 patients with pneumonia.
      • Marrie T.J.
      Pneumonia and carcinoma of the lung.
      • Holmberg H.
      • Kragsbjerg P.
      Association of pneumonia and lung cancer: the value of convalescent chest radiography and follow-up.
      • Søyseth V.
      • Benth J.S.
      • Stavem K.
      The association between hospitalisation for pneumonia and the diagnosis of lung cancer.
      A previous study that included older patients with 36% current smokers showed a 90-day lung cancer diagnosis rate of 2.48%.
      • Mortensen E.M.
      • Copeland L.A.
      • Pugh M.J.
      • et al.
      Diagnosis of pulmonary malignancy after hospitalization for pneumonia.
      Thus, our results are consistent with the literature.
      A strong correlation was found between pneumonia location and the risk of subsequent lung cancer. We believe this finding might be explained by the fact that upper-lobe pneumonia is significantly less common than lower-lobe pneumonia.
      • Rosenblatt M.B.
      • Bachman A.
      Upper lobe pneumonia in an adult.
      Thus, an equal number of new malignancies in upper and lower lobes will make upper-lobe pneumonia much more likely than lower-lobe pneumonia to represent an occult cancer. This hypothesis is supported by our finding that approximately 75.8% of lung cancers were located in the lobes involved by pneumonia, probably representing either postobstructive pneumonia or the lung cancer itself initially diagnosed as pneumonia.
      Lung cancer incidence decreased with time from pneumonia diagnosis. Lung cancer incidence after 365 days was similar to the incidence demonstrated in the NLST population.
      • Aberle D.R.
      • Adams A.M.
      • Berg C.D.
      • et al.
      The National Lung Screening Trial Research Team
      Reduced lung-cancer mortality with low-dose computed tomographic screening.
      This implies that heavy smokers hospitalized due to pneumonia are at an increased risk for lung cancer compared with heavy smokers without pneumonia for approximately 1 year. We were unable to determine the at-risk period more precisely due to the low number of patients involved. Non-lung cancer incidence was unrelated to pneumonia (P = .0001 for temporal relation of lung cancer vs non-lung cancer to pneumonia).
      Lung cancer histology distribution in our study was different from that of the general population.
      • Gandini S.
      • Botteri E.
      • Iodice S.
      • et al.
      Tobacco smoking and cancer: a meta-analysis.
      While 36.4% of new lung cancers were of small cell histology, considerably higher than expected, only 36.4% were adenocarcinomas. The most common non-lung malignancy diagnosed was head and neck squamous cell carcinoma, a relatively uncommon malignancy. These findings could be explained by the well-established relationship between these malignancies and heavy smoking.
      • Govindan R.
      • Page N.
      • Morgensztern D.
      • et al.
      Changing epidemiology of small-cell lung cancer in the United States over the last 30 years: analysis of the surveillance, epidemiologic, and end results database.
      • Lee P.N.
      • Forey B.A.
      • Coombs K.J.
      Systematic review with meta-analysis of the epidemiological evidence in the 1900s relating smoking to lung cancer.
      The 31 patients diagnosed with lung cancer in the first year following pneumonia could be divided into 3 groups. The first group includes patients with “background,” or expected cases of lung cancer diagnosed in heavy smokers, unrelated to the hospitalization due to pneumonia. On the basis of 1-year lung cancer incidence in the NLST (0.65%), we believe this group includes 3 patients (0.8%). The second group includes the 9 patients who were diagnosed with lung cancer within 30 days of pneumonia following clinical or radiological findings, pointing to a likely diagnosis of malignancy. The third group includes the 19 patients diagnosed more than a month after the pneumonia. For these patients the pneumonia was probably the first symptom of an occult lung cancer. Excluding the second group, cumulative lung cancer incidence in our cohort was 5.69%. These patients could probably have been diagnosed earlier if they had been screened for lung cancer with chest computed tomography.
      Most of the patients included in the study would be eligible for lung cancer screening with low-dose computed tomography according to the NLST criteria; only 76 (19.9%) were younger than 55 years of age, and all were heavy smokers. However, there is much controversy regarding this recommendation. Complication rates might be higher in “real world” patients, who are older and less healthy than trial volunteers, and who would be treated outside the high-volume medical centers that participated in the NLST. A higher rate of false-positive results and ensuing unnecessary biopsies, anxiety, and surgical procedures may adversely affect outcomes. The likely magnitude of lung cancer overdiagnosis resulting from wide implementation of screening is hard to assess. The long-term association of repeated low-dose computed tomography for heavy smokers with the development of radiation-induced cancers is unknown. Finally, cost-effectiveness issues have not been resolved.
      • Bach P.B.
      • Mirkin J.N.
      • Oliver T.K.
      • et al.
      Benefits and harms of CT screening for lung cancer: a systematic review.
      • Field J.K.
      • Oudkerk M.
      • Pedersen J.H.
      • Duffy S.W.
      Prospects for population screening and diagnosis of lung cancer.
      • Woolf S.H.
      • Harris R.P.
      • Campos-Outcalt D.
      Low-dose computed tomography screening for lung cancer: how strong is the evidence?.
      • Patz Jr., E.F.
      • Pinsky P.
      • Gatsonis C.
      • et al.
      NLST Overdiagnosis Manuscript Writing Team
      Overdiagnosis in low-dose computed tomography screening for lung cancer.
      • Black W.C.
      • Gareen I.F.
      • Soneji S.S.
      • et al.
      National Lung Screening Trial Research Team
      Cost-effectiveness of CT screening in the National Lung Screening Trial.
      Indeed, the Detection And screening of early lung cancer with Novel imaging TEchnology (DANTE) trial failed to demonstrate the benefit of low-dose computed tomography screening for lung cancer in heavy smokers, although perhaps this was due to its limited statistical power.
      • Kovalchik S.A.
      • Tammemagi M.
      • Berg C.D.
      • et al.
      Targeting of low-dose CT screening according to the risk of lung-cancer death.
      Several researchers have proposed that a more refined risk assessment could improve patient selection process and reduce the number of false-positive results, unnecessary complications, anxiety, and number of patients needed to scan to prevent one lung cancer-related death.
      • Infante M.
      • Cavuto S.
      • Lutman R.F.
      • et al.
      Long-term follow-up results of the DANTE Trial, a randomized study of lung cancer screening with spiral computed tomography.
      • Raji O.Y.
      • Duffy S.W.
      • Agbaje O.F.
      • et al.
      Predictive accuracy of the Liverpool Lung Project risk model for stratifying patients for computed tomography screening for lung cancer: a case-control and cohort validation study.
      • Tammemägi M.C.
      • Katki H.A.
      • Hocking W.G.
      • et al.
      Selection criteria for lung-cancer screening.
      • Pinsky P.F.
      • Berg C.D.
      Applying the National Lung Screening Trial eligibility criteria to the US population: what percent of the population and of incident lung cancers would be covered?.
      • Maisonneuve P.
      • Bagnardi V.
      • Bellomi M.
      • et al.
      Lung cancer risk prediction to select smokers for screening CT: a model based on the Italian COSMOS Trial.
      We believe we have identified a patient population, namely heavy smokers over 40 years old hospitalized due to community-acquired pneumonia, which is at an increased risk for lung cancer and might benefit from an early chest computed tomography scan. Lung cancer incidence in our study was considerably higher than in patients at the top quintile of lung cancer risk in the NLST data analysis.
      • Lee P.N.
      • Forey B.A.
      • Coombs K.J.
      Systematic review with meta-analysis of the epidemiological evidence in the 1900s relating smoking to lung cancer.
      The number needed to scan to detect a single case of new lung cancer in this population according to our data would be 12.3 (for cumulative incidence of 8.14%). The number needed to scan excluding patients diagnosed within 30 days of pneumonia, perhaps representing patients with overt lung cancer at pneumonia diagnosis, would be 17.6 (for cumulative incidence of 5.69%). We are unable to determine the ideal timing for the computed tomography due to the retrospective nature of our study. However, as most of the lung malignancies were probably present during the initial hospitalization, either as the direct cause of a postobstructive pneumonia, as the true etiology of the chest radiograph abnormalities initially diagnosed as pneumonia, or as a risk factor for severe pneumonia, we believe a low-dose chest computed tomography scan should be acquired during the hospitalization, to facilitate an early diagnosis. Additional computed tomography should be considered after 4-8 weeks when appropriate, when radiographic abnormalities due to pneumonia are expected to have resolved. It is unknown whether screening these patients with an early computed tomography scan will be translated to a prolonged survival. The high risk of lung cancer in these patients underlies the need for in-patient smoking cessation programs to utilize the hospitalization as an opportunity for the patient to quit smoking.
      Our study has several important limitations. The retrospective patient identification depends on accurate diagnosis coding. While we believe most pneumonia patients were correctly coded, it is likely that many smokers, and even more often former smokers, weren't coded adequately. This might have caused patient omission from the study cohort. Furthermore, it seems likely that the study's population is biased toward heavier smokers, who are probably more likely to be coded. A mean of 61 pack-years smoked and only 174 smokers excluded due to lack of heavy smoking seems to support this hypothesis. Therefore, our findings might be restricted to heavier smokers than intended, with lower cancer rates for patients with a smoking history of “only” 30 pack years.
      The retrospective nature of the study might cause bias due to unknown or unrecorded confounders. For example, family history of cancer was found to be the only factor significantly associated with lung cancer. As family history was documented for only 102 patients, the significant association found probably represents an increased tendency to document positive as opposed to negative family history. However, all studies assessing lung cancer risk after pneumonia were retrospective, and most relied on International Classification of Diseases coding analysis alone without chart review. As some pneumonia diagnoses are made inaccurately and smoking history is heterogeneous regarding number of pack-years, we believe our chart review is a major strength of our study.
      Data regarding malignancies were obtained from the INCR. All new cancers diagnosed are reported to the registry directly from the pathology institutions. Lung cancer patients might be missed if no biopsy was obtained, for example, in the case of patients deemed too ill for any treatment. However, such patients probably wouldn't be candidates for CT screening either.
      As this is a single-center study, it is more vulnerable to unknown bias. This is demonstrated by the high number of patients excluded due to medical history of pulmonary transplantation, due to our center being Israel's main lung transplant center. We collected all data that seemed relevant to lung cancer risk, but an unknown bias might exist. However, the single-center study design might contribute to data homogeneity, as patients were treated by the same physicians and chest radiographs reviewed by the same radiologists.
      Another major limitation is the lack of a comparison group. We chose to compare our results with published cancer incidence among heavy smokers and patients diagnosed with pneumonia rather than use a comparison group that would have been considerably smaller. While our inclusion criteria were similar to the NLST's, lung cancer rate was much higher in our study.
      In conclusion, 1-year lung cancer incidence of 8.14% (95% CI, 5.9%-11.2%) was found in a cohort of heavy smokers hospitalized due to community-acquired pneumonia. Patients with upper-lobe pneumonia were at a significantly higher risk for subsequent lung cancer of up to 23.8% (95% CI, 14.9%-40%). Seventy-five percent of lung cancers were found to involve the same lobes involved by the preceding pneumonia. Further evaluation with chest computed tomography should be strongly considered for these patients. A large prospective trial is needed to investigate whether early diagnosis of lung cancer in this patient population will translate to prolonged survival and lower rates of deaths attributed to lung cancer.

      Acknowledgment

      The authors thank Ms. Irena Liphshitz, Israel National Cancer Registry, for performing the linkage of the study population with the cancer registry database.

      References

        • Humphery L.L.
        • Deffebach M.
        • Pappas M.
        • et al.
        Screening for lung cancer with low dose computed tomography: a systemic review to update the US preventive task force recommendations.
        Ann Intern Med. 2013; 159: 411-420
        • Aberle D.R.
        • Adams A.M.
        • Berg C.D.
        • et al.
        • The National Lung Screening Trial Research Team
        Reduced lung-cancer mortality with low-dose computed tomographic screening.
        N Engl J Med. 2011; 365: 395-409
      1. U.S. Preventive Services Task Force (USPSTF). Final recommendation statement. Lung cancer: screening. Available at: http://www.uspreventiveservicestaskforce.org/Page/Document/RecommendationStatementFinal/lung-cancer-screening. Accessed September 2, 2015.

        • Centers for Medicare and Medicaid Services
        Decision memo for screening for lung cancer with low dose computed tomography.
        2015 (Available at:) (Accessed September 2, 2015)
        • Bach P.B.
        • Mirkin J.N.
        • Oliver T.K.
        • et al.
        Benefits and harms of CT screening for lung cancer: a systematic review.
        JAMA. 2012; 307 ([Erratum, JAMA. 2012;308:1324.]): 2418-2429
        • Field J.K.
        • Oudkerk M.
        • Pedersen J.H.
        • Duffy S.W.
        Prospects for population screening and diagnosis of lung cancer.
        Lancet. 2013; 382: 732-741
        • Woolf S.H.
        • Harris R.P.
        • Campos-Outcalt D.
        Low-dose computed tomography screening for lung cancer: how strong is the evidence?.
        JAMA Intern Med. 2014; 174: 2019-2022
        • Patz Jr., E.F.
        • Pinsky P.
        • Gatsonis C.
        • et al.
        • NLST Overdiagnosis Manuscript Writing Team
        Overdiagnosis in low-dose computed tomography screening for lung cancer.
        JAMA Intern Med. 2014; 174: 269-274
        • Black W.C.
        • Gareen I.F.
        • Soneji S.S.
        • et al.
        • National Lung Screening Trial Research Team
        Cost-effectiveness of CT screening in the National Lung Screening Trial.
        N Engl J Med. 2014; 371: 1793-1802
      2. Centers for Disease Control and Prevention (CDC). National Center for Health Statistics. Pneumonia. Available at: http://www.cdc.gov/nchs/fastats/pneumonia.htm. Accessed September 2, 2015.

        • Mandell L.A.
        • Marrie T.J.
        • Grossman R.F.
        • Chow A.W.
        • Hyland R.H.
        Canadian guidelines for the initial management of community-acquired pneumonia: an evidence-based update by the Canadian Infectious Diseases Society and the Canadian Thoracic Society. The Canadian Community-Acquired Pneumonia Working Group.
        Clin Infect Dis. 2000; 31: 383-421
        • Niederman M.S.
        • Mandell L.A.
        • Anzueto A.
        • et al.
        • American Thoracic Society
        Guidelines for the management of adults with community-acquired pneumonia. Diagnosis, assessment of severity, antimicrobial therapy, and prevention.
        Am J Respir Crit Care Med. 2001; 163: 1730-1754
        • Ramsdell J.
        • Narsavage G.L.
        • Fink J.B.
        • American College of Chest Physicians' Home Care Network Working Group
        Management of community-acquired pneumonia in the home: an American College of Chest Physicians clinical position statement.
        Chest. 2005; 127: 1752-1763
        • Lim W.S.
        • Baudouin S.V.
        • George R.C.
        • et al.
        • Pneumonia Guidelines Committee of the BTS Standards of Care Committee
        BTS guidelines for the management of community acquired pneumonia in adults: update 2009.
        Thorax. 2009; 64: 1-55
        • Tang K.L.
        • Eurich D.T.
        • Minhas-Sandhu J.K.
        • Marrie T.J.
        • Majumdar S.R.
        Incidence, correlates, and chest radiographic yield of new lung cancer diagnosis in 3398 patients with pneumonia.
        Arch Intern Med. 2011; 171: 1193-1198
        • Marrie T.J.
        Pneumonia and carcinoma of the lung.
        J Infect. 1994; 29: 45-52
        • Holmberg H.
        • Kragsbjerg P.
        Association of pneumonia and lung cancer: the value of convalescent chest radiography and follow-up.
        Scand J Infect Dis. 1993; 25: 93-100
        • Søyseth V.
        • Benth J.S.
        • Stavem K.
        The association between hospitalisation for pneumonia and the diagnosis of lung cancer.
        Lung Cancer. 2007; 57: 152-158
        • Mortensen E.M.
        • Copeland L.A.
        • Pugh M.J.
        • et al.
        Diagnosis of pulmonary malignancy after hospitalization for pneumonia.
        Am J Med. 2010; 123: 66-71
        • U.S. Department of Health and Human Services
        • Public Health Service
        • Office of the Surgeon General
        The Health Consequences of Smoking—50 Years of Progress. A Report of the Surgeon General 2014.
        (Executive Summary) U.S. Department of Health and Human Services, Rockville, MD2014 (Available at:) (Accessed September 2, 2015)
        • Almirall J.
        • Bolibar I.
        • Serra-Prat M.
        • et al.
        New evidence of risk factors for community-acquired pneumonia: a population-based study.
        Eur Respir J. 2008; 31: 1274-1284
        • Nuorti J.P.
        • Butler J.C.
        • Farley M.M.
        • et al.
        Cigarette smoking and invasive and invasive pneumococcal disease. Active Bacterial Core Surveillance Team.
        N Engl J Med. 2000; 342: 681-689
        • Nichols L.
        • Saunders R.
        • Knollmann F.D.
        Causes of death of patients with lung cancer.
        Arch Pathol Lab Med. 2012; 136: 1552-1557
      3. Centers for Disease Control and Prevention (CDC). CDC/NHSN surveillance definitions for specific types of infection. Available at: http://www.cdc.gov/nhsn/PDFs/pscManual/17pscNosInfDef_current.pdf. Accessed September 2, 2015.

        • Fine J.P.
        • Gray R.J.
        A proportional hazards model for the subdistribution of competing risk.
        J Am Stat Assoc. 1999; 466: 496-509
        • Rosenblatt M.B.
        • Bachman A.
        Upper lobe pneumonia in an adult.
        Ann Intern Med. 1938; 11: 1845-1855
        • Gandini S.
        • Botteri E.
        • Iodice S.
        • et al.
        Tobacco smoking and cancer: a meta-analysis.
        Int J Cancer. 2008; 122: 155-164
        • Govindan R.
        • Page N.
        • Morgensztern D.
        • et al.
        Changing epidemiology of small-cell lung cancer in the United States over the last 30 years: analysis of the surveillance, epidemiologic, and end results database.
        J Clin Oncol. 2006; 24: 4539-4544
        • Lee P.N.
        • Forey B.A.
        • Coombs K.J.
        Systematic review with meta-analysis of the epidemiological evidence in the 1900s relating smoking to lung cancer.
        BMC Cancer. 2012; 12: 385
        • Kovalchik S.A.
        • Tammemagi M.
        • Berg C.D.
        • et al.
        Targeting of low-dose CT screening according to the risk of lung-cancer death.
        N Engl J Med. 2013; 369: 245-254
        • Infante M.
        • Cavuto S.
        • Lutman R.F.
        • et al.
        Long-term follow-up results of the DANTE Trial, a randomized study of lung cancer screening with spiral computed tomography.
        Am J Respir Crit Care Med. 2015; 191: 1166-1175
        • Raji O.Y.
        • Duffy S.W.
        • Agbaje O.F.
        • et al.
        Predictive accuracy of the Liverpool Lung Project risk model for stratifying patients for computed tomography screening for lung cancer: a case-control and cohort validation study.
        Ann Intern Med. 2012; 157: 242-250
        • Tammemägi M.C.
        • Katki H.A.
        • Hocking W.G.
        • et al.
        Selection criteria for lung-cancer screening.
        N Engl J Med. 2013; 368: 728-736
        • Pinsky P.F.
        • Berg C.D.
        Applying the National Lung Screening Trial eligibility criteria to the US population: what percent of the population and of incident lung cancers would be covered?.
        J Med Screen. 2012; 19: 154-156
        • Maisonneuve P.
        • Bagnardi V.
        • Bellomi M.
        • et al.
        Lung cancer risk prediction to select smokers for screening CT: a model based on the Italian COSMOS Trial.
        Cancer Prev Res (Phila). 2011; 4: 1778-1789