Solitary pulmonary nodules (SPN) are a significant clinical problem and are predominantly detected incidentally on chest x-ray or CT. Traditionally, SPNs were nodules 1 to 3 cm in size. More than 75% of the patients are asymptomatic at the time of detection (6). The major clinical question is what is the etiology of this nodule, benign or malignant? This leads to asking if the SPN should be followed or resected. The evaluation becomes risk assessment, which becomes the “cost” of resecting a benign nodule versus the “cost” of missing a malignancy. Multidetector CT has started to make the “solitary” part of SPN a disappearing entity and adds complexity, as the SPN on chest x-ray becomes one of many nodules being detected, although most of the nodules are small (<1 cm). Thus, work-up of nodules less than 1 cm in size is more common than it once was.

Attacking the question of a pulmonary nodule has led to a wide variety of management strategies. All strategies clearly start with establishing the presence or absence of the nodule on a prior chest x-ray or CT; size stability over a 2-year time period is accepted as evidence of benignity. Unfortunately, it is often the case that no prior imaging studies are available, and the stability of lesion size cannot be confirmed. Otherwise, a three-pronged individual risk assessment is employed. Patient risk in terms of age, smoking, prior cancer history, occupational exposure to carcinogens, and the presence of hemoptysis are all weighed as well as the presence of any prior diagnosis of cancer. The pulmonary nodule evaluated by helical, preferably multidetector, CT is assessed for
nodule size, calcification, border regularity, density, and cavitation. The individual patient’s risk for tissue sampling and surgery is next weighed. This risk assessment requires significant mental calculus for the clinician, often facing discordant information (7). In general, the smaller the nodule, the younger the patient, the less the exposure to carcinogens, the greater the probability that the nodule is benign. However, risk of cancer often falls into an intermediate range, and a decision must be made as to whether to observe the lesion or to secure a histological diagnosis.

A variety of tissue sampling techniques are available: fiberoptic bronchoscopy and biopsy, percutaneous needle aspiration or needle biopsy, video-assisted thoracoscopy, and thoracotomy. Bronchoscopy is least invasive and thoracotomy will yield the most certain histologic opinion. The morbidity and small, but real, risk of death increase with the invasiveness of the tissue sampling procedure. This risk of invasiveness correlates with increased certainty of the yield of biopsy from about 50% to nearly 100% (8,9,10). The clinician commonly weighs watchful waiting with serial CT scans every 3 months for 2 years versus a tissue sampling procedure and then deciding which procedure. Nodules near the periphery of the lung are better suited to needle biopsy, whereas nodules centrally located near the major airways are often better candidates for bronchoscopic biopsy. Nodules in the middle area of the lung are often quite hard to reach reliably by either methods. Since the nonsurgical approaches have sensitivities less than 100%, a negative study does not exclude the presence of cancer. Thus, some centers often try to move to the most definitive procedure quickly, especially if a negative result will not be sufficient to plan future management.

An FDG PET has proved to be quite sensitive for the detection and evaluation of pulmonary nodules. Gould et al. (11), in a meta-analysis of 40 published studies, demonstrated that the sensitivity of FDG PET was 96.8% and the specificity was 77.8% in 1,474 focal lesions, all greater than 1 cm. FDG PET compared to histology in these published studies was highly sensitive and moderately specific. They stressed the negative predictive value of 97.6% of FDG PET. This suggested that applied to patient populations with 1 cm or larger nodules similar in risk factors to those from the meta-analysis, a patient with a negative FDG PET would have a posttest likelihood of malignancy in the SPN of less than 3% (Fig. 8.9.1). One of the challenges, however, is that the risk is not reduced to 0%, so that some follow-up imaging remains necessary if a choice is made to observe a patient after a negative PET study. Another challenge is that physicians are often asked to perform PET scans in patients with solitary pulmonary nodules less than 1 cm in diameter. In such patients, the diagnostic sensitivity is probably lower due to recovery coefficient limitations of PET scanners and the scan becomes of greatest value when positive.

Many schemas combining PET and biopsy strategies have been proposed. Typically, PET and biopsy provide complementary information (12). In some approaches, tissue sampling is used for greatest certainty, and PET covers those patients where the nodule was inconvenient to biopsy or nondiagnostic. In other schema PET is performed first, and if PET is positive, patients may in some centers go straight to surgery. Even when PET is applied only when the biopsy was nondiagnostic, FDG PET performs well to diagnose lung cancer (13) (Fig. 8.9.2).

Gould et al. (7) used decision analysis to create a management strategy for the physician managing the deployment of the many available diagnostic options for pulmonary nodules. Their analysis found four major roles for PET in evaluating a pulmonary nodule: (a) when a patient’s risk of lung cancer and CT findings are not congruent, (b) when the surgical or procedure biopsy risk is high, (c) when a nondiagnostic biopsy occurs, or (d) when the patient is uncomfortable with watchful waiting. The sensitivity analysis demonstrated the broad utility of PET for evaluating pulmonary nodules.

Dynamic contrast enhancement with multidetector CT (MDCT) is more effective in categorizing SPNs as benign or malignant than noncontrast CT. A recent evaluation of integrated PET/ CT was compared to dynamic contrast MDCT. FDG PET/CT had a significant advantage in sensitivity (96% vs. 81%) with similar specificity (88% vs. 93%). PET/CT exhibited negative predictive values similar to established PET data, but with improved positive predictive value. In a head-to-head comparison of state-of-the-art
imaging technology, PET/CT provides greater accuracy than MDTC. These techniques may be complementary, but the use of contrast-enhanced CT for characterization of solitary pulmonary nodules does not appear to be as widely adopted as is PET/CT scanning, which seems appropriate based on the performance characteristics of the two methodologies (14).

Integrated PET/CT compared to PET provides somewhat improved or similar nodule resolution, faster scanning times, and CT visualization of SPN. PET/CT technology provides additional methods of evaluating and improving evaluation of a pulmonary nodule. Dual point imaging involves FDG uptake measurements at two different times, typically separated by an hour. The SPN is evaluated by both its intensity of FDG uptake and the change over time. Most lung cancers have a high standardized uptake value (SUV) and the SUV increase is greater than 10% in the 1-hour period. Some, but not all, inflammatory processes have a decline in FDG uptake in this period of observation. However, some inflammatory processes can have rises in FDG uptake with time, and most centers do not currently apply the dual time point technique, in part due to logistical reasons (15). Although virtually all of the literature on PET and PET/CT with FDG has been developed using a 50- to 60-minute uptake period, there are some groups that perform the PET imaging at 90 minutes or more post-FDG injection. Although tumor/background ratios typically rise in this period of time, some caution must be used in interpretation, as interpretation criteria were developed based on the 1-hour uptake period and may differ slightly with a 90-minute uptake period.

The apparent FDG uptake of a nodule is impacted by the partial value effect. The size of the SPN on CT can be used to correct for partial volume effect. Initial research suggests that partial volume correction improves SUV measurements of nodules smaller than 2 cm and increases the sensitivity for malignancy (16). Movement of a SPN with respiration during a PET/CT study causes the counts from the FDG uptake to be distributed over multiple pixels. However, for small nodules, the correction in the SUV is substantial and dependent on lesion size. Measurements of lesion size for small lesions are quite subject to variability, and this variability can be propagated to the SUV assessments. Thus, many interpreting physicians perform a “visual” partial volume correction, in which they use their clinical judgment to realize that modest levels of FDG uptake, sometimes no more than blood pool, in lesions under 1 cm, may be indicative of the presence of a small cancer. Respiratory gating, especially in the lower lobes, corrects for the respiratory excursion of 1 to 3 cm during PET imaging. The apparent activity within the moving nodule may increase between 20% and 36% (17). These techniques show promise but need further investigation before full utilization is likely. It must be realized that some pulmonary nodules that are malignant or behave in a malignant fashion may have relatively low SUVs. Bronchioalveolar carcinoma is very commonly in this category and in the pure state they have SUV values that are much lower than other cancers, sometimes less than 2.5. Carcinoid tumors, both primary and metastatic, can also have low SUVs relative to more typical lung cancers. The CT characteristics of bronchioalveolar carcinoma often are quite suggestive of the presence of this tumor, even if the SUV is low.

When interpreting PET in SPN, a variety of interpretation schemes can be used. Since most PET imaging involves the use of PET/CT, it is important that the CT be examined to determine lesion size. Smaller lesions, less than two times the resolution of the imaging system, will not have all counts recovered, so their apparent SUV will be lower than their true radioactivity concentration. In many centers, a qualitative interpretation is applied in which SPNs with FDG uptake greater than that in blood pool background are called “positive” while those with less than blood pool background are called “negative.” Quantitative cutoff values of SUV have been proposed with 2.5 as a good separator of malignant and benign nodules; however, in practice, measurement errors could make a lesion with a true SUV of 2.7 appear to have an SUV of 2.4 (due to variance of the measurement method). Most centers tend to err on the side of sensitivity as opposed to specificity in interpreting PET of SPN to avoid false-negative results. A review of the meta-analysis of Gould et al. (7) shows that although a joint operating sensitivity/
specificity of 91% is possible with PET using FDG in SPN, most readers operate to achieve higher sensitivity at the tradeoff of lower specificity.

Non–small cell lung cancer (NSCLC) comprises about 80% of all lung cancers, and many patients are smokers or have had significant exposure to secondhand smoke. The most common cell types, in order of occurrence, are adenocarcinoma, including bronchoalveolar variants, squamous cell carcinoma, and large cell. In the past decade, the frequency of adenocarcinomas has grown steadily. Some of the patients with adenocarcinomas are nonsmokers, and a significant fraction are females (18). Surgical resection represents the best opportunity for cure, but recently combination therapy including chemotherapy, radiation therapy, and to some extent biological therapy has been employed with increasing frequency. NSCLC staging is based on the TNM system and requires accurate characterization of the primary tumor (T), regional lymph nodes (N), and distant metastases (M). Staging is important for appropriate selection of treatment and assessment of prognosis.