There is no universally accepted definition of a solitary pulmonary nodule (SPN). From a practical standpoint, the definition must be broad enough to not miss cancer but narrow enough to not waste resources and subject patients to unnecessary tests and procedures. From Gould and associates¹⁶ and the American College of Chest Physicians (ACCP) evidence-based clinical guidelines for lung nodules: “A solitary pulmonary nodule is a single, spherical, well-circumscribed, radiographic opacity that measures less than or equal to 3 centimeters in maximal diameter and is surrounded completely by aerated lung. There should not be any associated atelectasis, hilar enlargement or pleural effusion.” This definition does not specify the radiographic modality by which the lesion was detected or whether the patient had symptoms.

A solitary nodule that lacks radiographic characteristics to define it as either benign or malignant is an indeterminate nodule. An indeterminate nodule at presentation may be classified as benign after additional radiographic studies or with a review of all prior films. If the lesion cannot be defined as benign after further imaging, radiographic review, or biopsy, the lesion remains classified as indeterminate even if the suspicion of malignancy is low.

The prevalence of SPNs in the general population is unknown. The published rates depend on the manner of detection, the definition used in the study, and the population used in the study. Higher rates would be expected in populations that are older, those that include smokers, patients with known nonthoracic neoplasms, and patients who live in locations with high rates of mycobacterial or fungal infection. Owing to its superior resolution, CT is a more sensitive technique than plain radiography for identifying a pulmonary nodule.

Although the exact prevalence is unknown, SPNs are common clinical issues. An estimated 150,000 new solitary lung nodules are discovered each year.²⁴ Another group of patients are those at a higher risk for the development of lung cancer, and CT screening has been advocated for following these patients. In 1999, Henschke detected between one and three nodules on CT baseline screening in 7% of 1,000 patients who were >60 years of age and had a >10-pack-year smoking history.¹⁸ Twenty-seven of the 28 patients who underwent biopsy had lung cancer. Twenty-three of these resected nodules were stage 1 lung cancers. In a slightly different study, Swenson et al.⁴⁴ screened 1,520 patients above age 50 with at least a 20-pack-year smoking history and then screened the same patients again a year later.⁴ Sixty-six percent of the patients had at least one SPN on the later study. Twenty-five of these patients had lung cancer; 22 underwent operation for cure, and 7 patients had resection of benign nodules. More recently, MacRedmond and associates²⁸ identified 111 noncalcified nodules in 449 high-risk patients screened with serial chest CT over a 2-year period. Early lung cancer was found in 3 of these 111. Three other lung cancers were detected and were unresectable central tumors.

From these studies several points can be made. First, screening CT increases the number of patients with SPN who require attention. Second, even in the group most at risk, the majority of the lesions identified are benign. The false-positive rate is very high. Last, the cost of screening the general population at risk would be staggering. Mahadevia et al.²⁹ undertook a comprehensive analysis of the costs and clinical ramifications of screening CT. If half of the American population aged 45 to 75 who smoked underwent a single CT scan, 5 million SPNs would be identified. The cost was estimated to be $115 billion annually. As the patient population ages and direct patient advertising becomes more and more prevalent, SPNs will constitute a giant clinical issue.

The question once an SPN is identified is which of the nodules is cancer? Historically, the rates of malignancy in SPNs are quoted at 30% to 50%.^43,48 However, these rates come from older studies using chest x-ray as the screening method and did not include the use of CT as a screening tool. Geographical
location has clinical importance in the determination of malignancy rates. Of 137 consecutive nodules resected in the Midwest, 84% were benign and most were due to histoplasmosis.⁴⁹

Models for predicting the “pretest” probability of malignancy have been developed. Six independent predictors of malignancy were identified through multiple logistic regression analysis by Swenson et al.⁴⁶ These predictors are older age, current or past smoking, history of extrathoracic cancer >5 years before nodule detection, nodule diameter, spiculation, and upper lobe location. The same investigators found that the results of this model were similar to those of expert clinical judgment.⁴⁵ Interestingly, thoracoscopic rates of malignancy average between 48% and 60% from three thoracoscopic series.^3,9,27 These lower rates are likely due to the increased use of VATS as an effective, safe, and definitive method for the diagnosis of peripheral SPNs and to its eventual use, as opposed to long-term radiographic surveillance, in patients with lower clinical suspicion.

The odds of malignancy in an SPN that remains indeterminate after a careful evaluation and possible nonoperative biopsy is of vital importance to the general thoracic surgeon and should be used to formulate the clinical plan for the patient with an indeterminate nodule.

The differential diagnosis of an SPN encompasses an extensive list of diagnostic possibilities (Table 96-1). Lung cancer makes up 85% to 95% of malignant lesions and granulomas constitute a similar proportion of the benign group. All of the lung cancer subtypes may present as SPNs. Metastatic disease most commonly presents as multiple nodules but may manifest as a SPN. Metastatic disease is the next most common source of SPNs (10%–15%). Carcinoid tumors are rare and account for only 1% to 3% of solitary lesions. The other listed types of primary lung malignancies are exceedingly rare.

Hamartomas account for the largest proportion of benign tumors. In 1998, Fein,¹² in a meta-analysis of 3,802 tumors, noted that hamartoma accounted for 5% of tumors resected. The nonneoplastic benign nodules are most commonly due to granuloma from prior infections. In the United States, the fungal organisms Histoplasma capsulatum and Coccidiodes immitis are most common. In other parts of the world, prior Mycobacterium tuberculosis infections account for the majority of SPNs. Noninfectious etiologies typically occur as multiple lesions but occasionally present as a single density or dominant mass.

SPNs may be spurious as represented on radiographs. Extrapulmonary masses mimicking nodules include overlapping structures, nipples, soft tissue masses, and bony chest wall masses. CT scans, repeat radiographs, and/or markers are used to distinguish between spurious and real nodules.

A complete and thorough history and physical will contain helpful information to differentiate the diagnosis of a SPN. Cancer risk increases with older age, male gender, and a smoking history. Noninfectious nodules may be predicted by location and travel history. Midwesterners and southwesterners have granulomas more likely due to histoplasmosis and coccidiomyocosis, respectively. Most patients rarely recall the mild symptoms related to the primary infection. A history of exposure to mycobacteria or emigration from a country where tuberculosis is endemic will suggest tuberculosis as the etiology.

Symptoms may or may not correlate with the presence of an SPN. The symptoms should aid in the differential diagnosis of the lesion. Centrally located masses are more commonly associated with symptoms and peripheral lesions are often asymptomatic. Central masses are associated with dyspnea, wheezing, hemoptysis, pneumonia, and sputum production. Peripheral lesions may be associated with a dry, nonproductive cough and even vague chest pain. Nonmalignant infectious or inflammatory processes often produce systemic symptoms associated with acute pulmonary symptoms. Symptoms such as cough with purulent sputum, limited hemoptysis, dyspnea, malaise, fatigue, anorexia, and fever are linked with an acute or subacute infectious or inflammatory process. When presented with this clinical scenario, a reasonable course is to follow the patient with
radiographs for a short time prior to engaging in aggressive imaging or biopsy. Antibiotics are controversial and their use should be limited in duration, with appropriate follow-up imaging and reentry into the diagnostic algorithm in case the nodule fails to resolve.

Prior or synchronous malignancy must be considered when new solitary nodules are discovered. Multiple new nodules usually represent metastatic disease but may also represent reaction to chemotherapeutic agents or secondary infections from either the malignancy or its treatment. Multiple nodules assessed by CT scan usually represent metastases, although they may be benign if small. Johnson et al.²¹ noted that less than one-third of nodules <0.5 cm were metastatic disease. In 2001, Carucci and associates⁴ reported that clustered nodules on CT were not malignant. Clustered nodules were defined as geographically localized multiple nodules within 1 cm of others without a dominant nodule.

New nodules in patients with a history of nonpulmonary cancer increase the chances that the nodule is metastatic. The clinician must remember that even though the patient has a history of malignancy, the majority of solitary nodules are still due to benign disease or primary lung cancers. Two reports from 1987 and 1991 demonstrated that single pulmonary nodules in patients with known extrapulmonary malignancy are lung cancer more often than metastatic disease.^6,8 The probability that an SPN in a patient with a history of cancer is malignant depends on the type and stage of the initial cancer in addition to the individual’s risk of lung cancer or a benign disease.

The organ of origin of the primary cancer also helps to determine the risk of metastatic disease to the lung. Sarcoma and germ cell tumors have a high predilection for metastasizing to the lung. This high risk is due to the younger age of the patients and the lower risk of lung cancer in those patients, in addition to the specific cancer biology of the primary tumors.³⁵ In addition, solitary nodules in prior melanoma patients are usually metastatic. Conversely, head and neck tumors appear prior to or synchronously with lung cancers. The common risk factor for these patients and the tumors is smoking. Approximately 25% of head and neck patients develop additional primary tumors, mostly in the lungs.

Most of the more common extrapulmonary adenocarcinomas are associated with an intermediate risk of metastasis in an SPN. In a large series published in 2001 examining resected SPNs from patients with prior extrapulmonary cancer, Sortini et al.⁴² determined that 42% were metastatic disease, 32% were lung cancer, and 27% were benign. In patients with previous lung cancer, a new nodule is due to new lung cancer, either metastatic or metachronous in origin. In patients without a history of cancer and without signs or symptoms of an undiagnosed primary, the chance that a solitary nodule represents metastatic disease is very low.