Introduction

Solitary pulmonary nodules (SPNs) are defined as single spherical lesions <3 cm in diameter and completely surrounded by lung tissue without associated atelectasis or adenopathy [1, 2]. SPNs are commonly found by chest radiography and thoracic computed tomography (CT), with at least 1 in 500 chest radiograms revealing a nodule [3]. The discovery of such nodules is of specific concern because of primary malignancy found in approximately 35% of SPN cases and solitary metastases accounting for another 23% [46]. Patient and nodule characteristics are factors that influence physicians in choosing different approaches, invasive and/or noninvasive, to diagnose a pulmonary nodule. Although early detection of malignant nodules could improve the prognosis of bronchogenic carcinoma, resection of a benign nodule might lead to mortality or morbidity [79]. Thus, the diagnosis and management of an SPN might have an impact on patients’ quality of life.

Many studies have investigated the quality of life of patients with lung cancer and its determinant factors [1013]. More recently, researchers reported on the prognostic role of quality of life in patient survival [1416]. In contrast, no study has focused on the quality of life of patients with SPNs. With more people being screened for lung cancer because of the increasing use of positron emission tomography (PET) and CT, a growing number of people will be diagnosed with a nonmalignant pulmonary nodule. Therefore, living with a pulmonary nodule may, to some extent, be a matter of concern for patients over time. In addition, the benefits for mortality of early lung cancer detection through screening are uncertain, and thus patient quality of life may be an outcome of interest for healthcare professionals.

Health-related quality of life (HRQoL) has been well recognized for several decades as an issue concerning the functional status and well-being of people [17, 18]. In general, the concept relates to subjective perceptions of the physical, mental and social health-related dimensions. HRQoL has become an important target in medical care, especially for assessing treatment outcomes in chronic diseases, and an important outcome criterion in randomized clinical trials, especially of oncology [19]. HRQoL information could also be useful in the interactions of physicians and patients in clinical practice, especially when they have to choose among treatments with equivalent effectiveness [20] in health services evaluation, in research and in health policy making.

Our study aimed to describe the HRQoL of patients with SPNs, both malignant and nonmalignant, to compare this HRQoL with the HRQoL of a sample of the French general population and to identify variables that might influence the HRQoL of patients with SPNs.

Materials and methods

We studied the HRQoL of patients who participated in the SPN diagnosis management study [7, 21], which aimed to evaluate the management of SPN diagnosis before (2002–2003) and after (2004–2005) the introduction of PET in the northeastern region of France. This prospective multicentre study involved five regions in the northeastern region of France (Alsace, Lorraine, Franche-Comté, Bourgogne and Champagne-Ardenne) with 18 health administrative districts covering 8,220,000 inhabitants [22]. The study was approved by the institutional review board (Commission Nationale Informatique et Liberté [CNIL]) in January 2002. We collected data on the before-PET introduction period, between May 2002 and March 2003, and the after-PET introduction period, between June 2004 and June 2005.

Study samples

We identified new SPN cases from CT reports that mentioned “spherical lesion” or “nodule” or synonymous terms (including “mass” or “solitary mass”, “solitary pulmonary nodule” or “SPN”). Patients were ≥18 years old, had an SPN between 1 and 3 cm in diameter, did not have a previous cancer and continued the diagnostic management in all care centres involved [7, 21]. We included only patients with confirmed SPN; 152 patients were included in 2002–2003 and 307 in 2004–2005 (Fig. 1). We excluded from the analysis data for patients who died during the diagnosis process and follow-up (n = 34); were unable to respond to our questionnaire, because either they did not understand the questionnaire or French (n = 66); too ill to respond, did not return the questionnaire, did not respond on time or did not complete it (n = 188). In our analysis, we considered that patients with indeterminate nodule at 6-month after the end of the diagnosis process as patients with a nonmalignant nodule. We did not look any further whether the nodule turned out to be a malignant nodule 2 years later.

Fig. 1
figure 1

Flow chart of patients through the trial of health-related quality of life of patients with single pulmonary nodules (SPNs)

For a comparison group, we took a sample from the decennial survey on population health and medical services involving 25,000 families in France. The Institute National de la Statistique et des Études Économiques (INSEE) conducted the survey between October 2002 and October 2003. Its main objective was to evaluate subjects’ perspectives and behaviour regarding their health. The survey involved an estimate of annual health resources consumption, use of physicians’ services, purchase of medicines prescribed and nonprescribed, as well as an investigation of subjects’ behaviour towards medical prevention, diet and exposure to risk, in particular, professional risks [23]. Among this population, 20,574 subjects completed the SF-36 questionnaire. The mean age of subjects was 45 years, and 52.9% of the sample was women. For our study, we selected 17,550 subjects of both sexes of a similar age range as that of our patients (27–90 years).

Data collection

HRQoL was assessed by a generic self-administered questionnaire at 6 months after the end of the diagnostic procedure. Patient quality of life reflected (1) the effect of the diagnosis of lung cancer and the beginning of treatment for patients who received a diagnosis of lung cancer or (2) the effect of simple surveillance for patients with nonmalignant nodules.

The generic instrument was the original version of Medical Outcomes Study Short Form 36 (SF-36), a multi-purpose health survey with 36 questions that measures HRQoL status during the previous 4 weeks. Scores range from 0 to 100, and high scores reflect better quality of life. The survey yields scores for 8 scales or dimensions of functional health and well-being: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role emotional and mental health. The questionnaire allows for a psychometrically based physical-component summary that measures the absence of physical limitations, disability or decrease in well-being and energy level and a mental-component summary that measures the absence of psychological distress and limitations in usual social/role activities because of emotional problems. This instrument has been useful in surveys of general and specific populations, comparing the relative burden of diseases and differentiating the health benefits produced by a wide range of treatments [24, 25].

Other measures

Other information collected were sociodemographic characteristics (age, sex, smoking history), nodule characteristics (spiculation, calcification and mediastinal node involvement), examinations and treatments prescribed during 6-month follow-up.

Statistical analysis

All variables were described for the entire sample and separately for patients with malignant and nonmalignant nodules in terms of percentages, means and standard deviations (SDs). Chi-square or Fisher exact tests and Mann–Whitney or Student t-tests were used to compare qualitative and quantitative variables, respectively.

Scores for the SF-36 were first compared between patients with malignant versus nonmalignant nodules and then between all SPN patients and the sample from the French general population {with similar age range (27–90 years)} by Student t-test. Each dimension of the SF-36 was then treated as the dependent variable in a multivariate linear regression model. We tested whether the HRQoL of patients with SPN differed from that in the general population sample. We adjusted the model on age, sex and smoking history and calculated the mean scores by Bonferroni correction. We tested first-order interaction of the group variable (patients with malignant or nonmalignant nodules or the general population) by age, sex and smoking history on SF-36 dimension scores.

We considered that a mean difference of at least 5 points of SF-36 scores as clinically meaningful [26].

A P < 0.05 was considered statistically significant. Data were recorded in a Microsoft Access database, and statistical tests involved use of SAS v9.1 (SAS Inst., Cary, NC).

Results

Patient characteristics

Of 171 patients (47.6%) participating in the study, 39 (22.8%) received a diagnosis of a malignant nodule at the end of the diagnostic procedure. Compared to patients who did not participate (n = 188), there was no difference in mean age (62.4 ± 13.1 vs. 64.3 ± 15.4 years; P = 0.2). Participating patients were more frequently men (73.1 vs. 64.4%; P = 0.06) and smokers (66.1 vs. 46.3%; P < 0.01) and more often consulted physician-specialists (88.3 vs. 80.8%; P = 0.05); the frequency of diagnosis of malignant nodules did not differ between groups (Table 1).

Table 1 Description of patients with solitary pulmonary nodules (SPNs), nodule and diagnosis process by patients who did and did not participate in the study 6 months after their diagnosis

Patients with malignant and nonmalignant SPNs did not differ in mean age (63.9 ± 9.8 vs. 62.7 ± 13.6; P = 0.8). Patients with malignant SPNs were more likely to be smokers (84.6%; P = 0.005)), had noncalcified SPNs (94.9%; P = 0.01) and spiculated SPNs (41.1%; P = 0.003) than those with nonmalignant SPNs (61.4, 77.3, 18.2%, respectively). Before receiving a diagnosis, patients with malignant SPNs underwent more diagnostic tests than did those with nonmalignant SPNs (mean 5.7 ± 4.4 vs. 2.7 ± 1.9; P < 0.01) (Table 2). More than 90% of patients underwent CT scan of the thorax, and about 40% had also bronchial fibroscopy [21].

Table 2 Characteristics of patients with SPNs, nodule and diagnosis process by diagnosis at 6 months after diagnosis

HRQoL measured by the SF-36

Patients with malignant SPNs had significantly lower HRQoL scores, by 5–14 points, than those with nonmalignant SPNs for 3 of 8 SF-36 dimensions—physical role (P = 0.04), emotional role (P = 0.02) and vitality (P = 0.04)—and lower mean scores for mental health (P = 0.06) and social functioning (P = 0.05) (Table 3).

Table 3 Comparison of health-related quality of life on the SF-36 by SPN diagnosis and for patients with SPNs and a sample of the French general population

For patients with SPNs, the highest mean HRQoL score was for the social functioning dimension (64 ± 27.1) and the lowest was for vitality (41.9 ± 20.5). Compared to the sample of the French general population, patients with SPNs had a significantly lower HRQoL (−11 to −30 points) in all dimensions (P < 0.001) on univariate analysis. Patients with nonmalignant SPNs had lower HRQoL scores (−9 to −26 points; P < 0.001) than those for the general population (Table 3).

On Bonferroni correction, the sample of the French general population showed significantly higher mean scores for all HRQoL dimensions than did patients with malignant SPNs (+6 to +29 points) and nonmalignant SPNs (+11 to +44 points). Meanwhile, patients with malignant SPNs had significantly lower scores for physical and emotional-role dimensions (−14 and −18 points, respectively) than those with nonmalignant SPNs (Fig. 2).

Fig. 2
figure 2

Mean HRQoL scores differences between groups for dimensions of the SF-36.

Note: * P < 0.05 (Bonferroni correction), Delta QoL the difference of mean scores between groups, GP general population

Factors related to HRQoL

On multivariate regression analysis, after adjusting for age, sex and smoking history, having an SPN was associated with low mean scores in all HRQoL dimensions (P < 0.001). Male sex was associated with high mean scores for all dimensions (P < 0.001), and older age and smoking history (smoker/ex-smoker) were related to low mean scores for all dimensions (P ≤ 0.001) (Table 4a, b). The results did not differ by group.

Table 4 Factors contributing to mean scores for health-related quality of life dimensions of the SF-36 for patients with SPN at 6 months after diagnosis

Discussion

This study quantifies how patients with SPNs perceive their HRQoL by use of the SF-36 and compares HRQoL with that of a sample of the French general population. All patients with SPNs felt tired and worn out (low scores for vitality), but regardless of their fatigue, patients could perform normal social activities without interference from physical or emotional problems (high score on social functioning). In addition, patients with malignant nodules had a lower HRQoL than did those with nonmalignant nodules, and patients with nonmalignant nodules had lower HRQoL scores than did people in the French population. Overall, having an SPN, regardless of its malignant nature, and older age were related to low HRQoL in all of SF-36 dimensions. As in many studies about quality of life, there was little variance explained in the model. It means that these characteristics alone could not explain the QoL of patients with SPN. Perhaps there were other variables that were not collected that influence these patients QoL. Therefore, further studies may be warranted in order to identify those factors.

A study in Italy found that the mean scores of physical and emotional role, bodily pain and physical composite scale (on SF-36) of patients with lung cancer, measured at 1 month after surgery were decreased from baseline. Then at 3 months, the scores increased to reach almost the baseline scores (P < 0.001 to P = 0.03) [27]. It means that QoL measured at 3 months after a surgery may be less affected than if measured at 1 month. In the United States, Handy Jr et al. showed that at 6 months after surgery, patients with lung cancer had significantly lesser mean scores of SF-36 in physical-, role-, social-functioning and bodily pain (P = 0.002 to P < 0.001) compared to the general population [28]. In our study, where the QoL was measured at 6 months after the end of the diagnostic process, first, we observed differences in mean scores of patients with malignant SPN versus those with nonmalignant SPNs, especially on physical role, emotional role and vitality (P = 0.04 to 0.02). These differences are certainly due to something that differentiates those 2 groups of patients. Table 2 showed that patients with malignant SPN were more likely to be smokers; underwent more diagnostics tests and received some treatments. Therefore, we think that these characteristics and also the nodule’s related condition were related to patients’ QoL. Second, the QoL of patients with nonmalignant SPN, where no treatment was administered, was also significantly different from that of the general population. Third, patients in both groups had lower mean scores than the general population in all of SF-36 dimensions.

Some limitations should be considered in applying the results of our study to other populations.

First, perhaps it is true that these patients had already a worse QoL in the first place compared to the general population, independently of their diagnoses and therapeutic procedures. Unfortunately, the design of our study could not confirm this hypothesis. Second, smoking history may relate to certain disease such as emphysema and thus may influence patients’ quality of life. As showed by Alonso et al., patients with chronic lung disease (including emphysema) had lesser mean scores (from −2 to −13 points) than those without this disease in all of SF-36 dimensions [29]. In this study, participating patients were more likely to be smokers/ex-smokers and, thus, might have other comorbidities, such as emphysema. For this reason, perhaps they had worse quality of life than those who did not participate. On the other hand, mediastinal involvement normally indicates the stage N of lung cancer [3032]. As showed on Table 1, nonparticipating patients were more likely to have a mediastinal involvement and, thus, might have a worse quality of life than those who did participate. It means that with a difference of −8 to −32 points between participating patients with the general population, it seemed that an SPN condition played a role in the alteration of QoL of these patients. Third, some patients may have been too ill to respond or simply were not willing to participate to this study. Fourth, we did not include data for some patients who responded after the 6 months of follow-up. Therefore, the result of the present study may still over-estimate the quality of life of patients with SPN.

Patients’ quality of life is important, whatever the condition is, because the concept can also reflect the relation among health practitioners, patients and the condition management process. However, little literature exists on the quality of life of patients with SPN. We found that the mean scores for patients with SPNs were significantly lower (by 11–30 points) in all dimensions of the SF-36 than those for a sample of the French general population (Table 4). The interpretation of these findings may still be problematic clinically [33] and depends on the perspective from which the change is viewed. From the patient perspective, a 5-point decline in QoL may represent enough an impact on an individual that it prompts the patient to seek medical treatment or alternative therapies designed to improve QoL [26]. In our study, for example, the largest differences in mean scores between the general population and our patients were observed in physical- and emotional-role functioning, as well as physical functioning (24, 30 and 32 points, respectively). It showed that patients with an SPN had difficulties in performing daily activities, personally or professionally, due to their physical health and emotional problems. Therefore, they might need medical or alternative interventions to overcome their problems as well as to increase their QoL. In other words, a meaningful change in HRQoL may be one that results in a meaningful reduction of symptoms or improvement in function [34].

This study reflected the extent to which the quality of life of patients with SPN at 6 months after the diagnosis process differed from that of the general population. Therefore, although people may not consider pulmonary nodules as obstacles to life, patients with such a nodule had lower quality of life than the general population. Our study contributes to understanding the QoL of patients with a diagnosed SPN and reveals possible targets for helping such patients to improve their quality of life. Clinicians may need to follow up their patients, provide counselling services and pay more attention to patients with a high probability of having malignant nodules.

As known, the recommended follow-up for patients with indeterminate nodules is a “wait-and-watch” strategy, with CT performed every 3 months for 2 years [1, 2, 3537]. Future work could investigate the quality of life of patients with indeterminate SPN or nonmalignant nodules during a 2-year follow-up process to evaluate the evolution of patient’s quality of life over time. As well, a qualitative approach could add information that cannot be assessed quantitatively. This approach may be useful for health practitioners to help patients with SPN cope with their disease and improve their HRQoL.