Nodulo pulmonar solitario

Aca les mando los criterios de valuacion de nódulo pulmonar solitario Diagnostic evaluation and initial management of the solitary pulmonary nodule Author Steven E Weinberger, MD Section Editors Nestor Muller, MDTalmadge E King, Jr, MDJames R Jett, MD Deputy Editor Kevin C Wilson, MD Last literature review version 16.3: octubre 2008 This topic last updated: enero 4, 2008 (More) INTRODUCTION — A solitary pulmonary nodule (SPN), or "coin lesion", is a common clinical problem [1]. It is usually detected incidentally on a plain chest radiograph or computed tomographic (CT) scan [2]. The major question that follows detection of a SPN is whether the lesion may be malignant, with management varying accordingly. In this topic review, a SPN is defined and its differential diagnosis is summarized. A general approach is then outlined, followed by detailed descriptions of the diagnostic evaluation and initial management. Radiographic evaluation of pulmonary nodules is discussed in greater detail separately. (See "Computed tomographic and positron emission tomographic scanning of pulmonary nodules"). DEFINITION — A SPN is a lesion that is both within and surrounded by pulmonary parenchyma. The size at which a "nodule" becomes a "mass" is arbitrary, although 3 or 4 cm is typically used [3,4]. DIFFERENTIAL DIAGNOSIS — The causes of a SPN can be categorized as benign or malignant (show table 1). The estimated frequency of each etiology varies substantially among studies, reflecting differences in the population studied, the methodology used, and the age of the study [4-10]. As an example, a higher prevalence of malignant etiologies among resected nodules is generally reported by newer studies, likely due to improved preoperative prediction of which nodules are malignant and require resection. Malignant etiologies — Common causes of a malignant SPN include primary lung cancer, carcinoid tumors, and lung metastases. Primary lung cancer — All types of primary lung cancer can present as a SPN. Adenocarcinoma (including the subtype bronchioloalveolar cell carcinoma) is the histologic subtype that most commonly presents as a SPN, followed by squamous cell carcinoma and large cell carcinoma. Both adenocarcinoma and large cell carcinoma share a tendency to originate as a peripheral lesion, whereas squamous cell carcinoma presents more frequently as a central lesion than as a peripheral nodule. Rarely, malignant lymphomas can present as a SPN. Carcinoid tumors — Carcinoid tumors tend to be centrally located endobronchial lesions; however, approximately 20 percent arise peripherally and present as a SPN. These tumors are generally well circumscribed radiographically. (See "Bronchial carcinoid tumors"). Metastatic cancer — Extrapulmonary malignancies that are most likely to produce a SPN due to metastasis include malignant melanoma, sarcomas, and carcinomas of the colon, breast, kidney, and testicle [11]. The probability of metastasis is approximately 25 percent when a SPN is detected on a chest radiograph in a patient with a history of extrathoracic malignancy [12]. Most metastases present as multiple pulmonary nodules. Benign etiologies — Common causes of a benign SPN include infectious granulomas and hamartomas. Less common causes include pulmonary abscesses, vasculitic lesions (as in Wegener's granulomatosis), and pulmonary dirofilariasis. Infectious Granulomas — Infectious granulomas cause approximately 80 percent of benign nodules [2,8,9]. Endemic fungi (eg, histoplasmosis, coccidioidomycosis) and mycobacteria (either tuberculous or nontuberculous mycobacteria) are the most frequently recognized causes of infectious granulomas presenting as a SPN. When nontuberculous mycobacterial disease presents as a solitary peripheral nodule, the cause of the nodule is typically unrecognized until the lesion is resected as a presumed primary lung cancer [13]. In patients with acquired immunodeficiency syndrome, pneumocystis jirovecii (previously called pneumocystis carinii) infection can present as a SPN and may cavitate [14]. (See "Clinical presentation and diagnosis of Pneumocystis carinii (P. jirovecii) infection in HIV-infected patients"). Hamartomas — Hamartomas are generally considered benign tumors of the lung. They cause approximately 10 percent of benign nodules [2,8,9]. Hamartomas typically present in middle age, grow slowly over years, and are histologically heterogeneous. Cartilage (with scattered calcification), fat, muscle, myxomatous tissue, and fibroblastic tissue may all exist (show picture 1, show histology 1A-1B) [15]. The characteristic appearance of a hamartoma on a chest radiograph is a SPN with "popcorn" calcification, although this pattern is observed in less than 10 percent of cases (show radiograph 1). High resolution CT scanning of the lesion is particularly useful because it may demonstrate focal areas of fat, or calcification alternating with fat, which are virtually diagnostic of a hamartoma [16]. Dirofilariasis — Pulmonary infestation with the dog heartworm, Dirofilaria immitis, is a rare but well-recognized cause of a SPN [17,18]. Dogs are the usual animal host, but cats, wolves, coyotes, and foxes can also harbor the organism. The greatest concentration of human cases in the United States is found in eastern, southeastern, and southern coastal states. (See "Miscellaneous nematodes"). The organism is transmitted to humans by Aedes, Anopheles, Culex, or Myzorhynchus mosquitoes, which ingest blood-containing microfilariae from affected dogs [19]. Larvae in humans embolize to the lungs (the larvae are dead because the human is not a suitable host for the organism), where antigens from the dead organism initiate endarteritis and a granulomatous response. The typical finding with Dirofilaria immitis affecting the lungs is a noncalcified, pleural-based nodule [17]. Pulmonary dirofilariasis can be indistinguishable from other causes of a SPN; therefore, it is often resected as a presumed primary lung cancer. Once the diagnosis is made, no further therapy is indicated. GENERAL APPROACH — The ideal approach to a SPN would result in definitive resection of all malignant nodules, while avoiding resection of all benign nodules that do not require therapy. Unfortunately, a foolproof means to implement such a strategy does not exist. Approaches that remove most malignant nodules tend to result in frequent removal of benign nodules, whereas approaches that leave most benign nodules intact tend to leave some malignant nodules unresected. The former approach is generally preferred because: The five year survival rate may be as high as 70 to 80 percent for patients who have had a stage 1A (T1N0M0) non-small cell lung cancer resected [20-23]. In contrast, unresected lung cancer is often fatal. The operative mortality rate is less than 1 percent among patients with benign nodules early stage lung cancer [2,24]. Patients who have a SPN undergo an initial diagnostic evaluation. During the evaluation, clinical features, radiographic features, and, occasionally, quantitative models are used to determine the probability that the nodule is malignant. Once the probability that the SPN is malignant has been formulated, initial management must be chosen. There are many reasonable approaches to the initial management of a SPN and significant variation exists among institutions. When decisions are uncertain, initial management should carefully consider the patient's preference [25,26]. DIAGNOSTIC EVALUATION — In this section, the clinical features, radiographic features, and quantitative models used to determine the likelihood that a nodule is malignant are discussed. Clinical features — Clinical features associated with an increased probability that a SPN is malignant include advanced patient age and underlying risk factors. Patient age — The probability of a SPN being malignant rises with increasing patient age [5,7,27]. As an example, one study stratified the percentage of SPNs that were malignant according to the patient's age [7]: - Three percent in patients between ages 35 and 39 - 15 percent between ages 40 and 49 - 43 percent between ages 50 and 59 - 50 percent or higher at age 60 or above Another study similarly found malignant nodules more frequently in patients 50 years of age or older, compared to patients less than 50 years of age (65 versus 33 percent) [5]. Despite the association between age and malignancy, a SPN should not be automatically assumed to be benign in a young patient [27]. Risk factors — The possibility of lung cancer is always a concern when a SPN is detected in a patient with a history of smoking, because of the strong association between cigarette smoking and primary lung cancer. Other risk factors for lung cancer (eg, asbestos exposure, family history) should also be considered when evaluating a patient with a SPN. Previously diagnosed malignancy increases the likelihood that a SPN is a metastasis [28]. (See "Overview of the risk factors, pathology, and clinical manifestations of lung cancer", section on Risk factors and See "Cigarette smoking and other risk factors for lung cancer"). Radiographic features — Radiographic features that can be used to predict whether a nodule is malignant include size, border, calcification, density, growth, and metabolic activity. Size — Larger lesions are more likely to be malignant than smaller lesions. In one study, the likelihood of malignancy was 0.2 percent for nodules smaller than 3 mm, 0.9 percent for nodules 4 to 7 mm, 18 percent for nodules 8 to 20 mm, and 50 percent for nodules larger than 20 mm [29,30]. (See "Computed tomographic and positron emission tomographic scanning of pulmonary nodules", section on Size). Border — Malignant lesions tend to have more irregular and spiculated borders, whereas benign lesions often have a relatively smooth and discrete border. (See "Computed tomographic and positron emission tomographic scanning of pulmonary nodules", section on Border). Calcification — The presence of calcification does not exclude malignancy. As an example, "eccentric" calcification (ie, asymmetric calcification) should raise concern about carcinoma arising in an old granulomatous lesion (ie, a "scar" carcinoma). Certain patterns of calcification, however, strongly suggest that a SPN is likely benign. These patterns include "popcorn" calcification (show radiograph 1), laminated (concentric) calcification, central calcification (show radiograph 2), and diffuse, homogeneous calcification (show radiograph 3). (See "Computed tomographic and positron emission tomographic scanning of pulmonary nodules", section on Calcification). Density — Increased density of a SPN argues against malignancy. As an example, one study measured the CT density of 58 malignant nodules and 33 benign nodules [31]. All of the malignant nodules were less than 147 Hounsfield units and most of the benign nodules were greater than 164 Hounsfield units. The likelihood of incorrectly identifying a malignant nodule as benign decreases if a higher cutoff value is used, although benign nodules then become more likely to be characterized as indeterminant [32]. Although density measurement was at one time considered to be a promising technique, it is no longer used as part of the routine evaluation of a SPN. (See "Computed tomographic and positron emission tomographic scanning of pulmonary nodules", section on Density). Growth — Review of available prior imaging studies is a critical part of the diagnostic evaluation. Lesions that are malignant tend to have a volume doubling time between 20 and 400 days, where volume doubling of a nodule corresponds to a 30 percent increase in its diameter [2,4,22,33]. Therefore, a SPN whose size has increased very rapidly or has remained stable for a prolonged duration is likely benign. Volume doubling time also varies according to the CT appearance of malignant SPNs. As an example, nodules described as ground glass, ground glass with a solid component, or solid had volume doubling times of 813, 457, and 149 days, respectively, in one study [29,34]. (See "Computed tomographic and positron emission tomographic scanning of pulmonary nodules", section on Growth). Quantitative models — Quantitative models (based upon Bayes' theorem) use likelihood ratios (LRs) to estimate the probability that a nodule is malignant [2,35]. LRs are defined as the prevalence of a specific finding among all subjects with a malignant nodule, divided by the prevalence of the same finding among all patients with a benign nodule. To estimate the probability of malignancy, LRs are first determined for nodule size, patient age, smoking history, and overall prevalence of malignancy in the population (show table 2) [36]. The odds of malignancy are then calculated by multiplying the LRs (show table 3). Finally, the probability of malignancy is calculated from the odds of malignancy (show table 3). This method of estimating the probability of malignancy was subsequently modified to include more LRs, thereby accounting for more radiographic and clinical characteristics (show table 4) [37].An alternative method of estimating the probability of malignancy has also been reported (show table 5) [38,39]. INITIAL MANAGEMENT — An initial management decision should be made once the probability that the SPN is malignant has been assessed. There is significant variation among institutions regarding what is considered optimal initial management of a SPN; one reasonable approach is as follows (show figure 1): A nodule that has a low probability of being malignant can be followed with serial CT scans. A nodule that is 1 cm or larger and has an intermediate probability of being malignant should be evaluated by FDG-PET. Nodules that are negative on FDG-PET can be followed with serial CT scans, while nodules that are positive should be excised. Sampling of the nodule should be performed if FDG-PET is unavailable. A nodule that is smaller than 1 cm and has an intermediate probability of being malignant can be followed by serial CT scans. A nodule that has a high probability of being malignant should be excised. It must be emphasized that this is just one of many reasonable approaches to the initial management of a SPN. Optimal management is individualized and carefully considers the patient's preference. Serial CT scans — Traditionally, a nodule that remained stable for two years or longer on a chest radiograph was considered likely benign. Several arguments have been used to challenge this practice, including: The two year cutoff is arbitrary. A small retrospective study reported that lack of appreciable growth on a chest radiograph over a two year duration had a positive predictive value of only 65 percent for a benign lesion [2,40,41]. In addition, certain cell types (bronchioloalveolar carcinoma and carcinoid) can maintain a stable size for two years or longer [2]. CT scans are preferable to chest radiographs. Growth is more readily detected by CT than chest radiography. Specifically, 3 to 5 mm of growth is necessary for detection on a chest radiograph, compared to only 0.5 mm on a high-resolution CT scan [2]. Different strategies should be used for nodules of different sizes. Nearly 50 percent of smokers older than 50 years have at least one nodule [29,42]. Moreover, approximately 10 percent of these patients develop new nodules over a one year period. The vast majority of these nodules are small and unlikely to be malignant [29,30,42]. Thus, the traditional approach of performing CT scans every three months during the first year and every six months during the second year is both unnecessary and impractical. The following guidelines, proposed by the Fleischner Society, advocate different frequencies of CT scans based upon the size of the nodule and the patient's risk for lung cancer (show table 6) [29]. Patients are considered low risk if they have no history or minimal history of smoking and other risk factors; otherwise, patients are considered high risk. Times refer to the duration from initial detection of the SPN: For nodules ≤4 mm, serial CT scans are not required if the patient is low risk. Patients who are high risk should have a CT scan performed at 12 months with no further follow-up if the nodule is unchanged. For nodules 4 to 6 mm, a CT scan should be performed at 12 months if the patient is low risk, with no further follow-up if the nodule is unchanged. Patients who are high risk should have a CT scan performed at six to 12 months and at 18 to 24 months if the nodule is unchanged. For nodules 6 to 8 mm, a CT scan should be performed at six to 12 months and at 18 to 24 months if the nodule is unchanged and the patient is low risk. Patients who are high risk should have a CT scan performed at three to six months, nine to 12 months, and 24 months if the nodule remains unchanged. For nodules greater than 8 mm, a CT scan should be performed at three, nine, and 24 months if the nodule remains unchanged, regardless of whether the patient is low or high risk. These guidelines are not applicable to young patients (less than 35 years old), or patients with an extrathoracic malignancy or unexplained fever [29]. Metabolic imaging — 18-flourodeoxyglucose positron emission tomography (FDG-PET) can help distinguish malignant and benign lesions because cancers are metabolically active and take up FDG avidly. It is indicated for patients whose SPN is 1 cm or greater in size and has an intermediate probability of malignancy, especially if the patient has increased surgical risk. (See "Computed tomographic and positron emission tomographic scanning of pulmonary nodules", section on Positron emission tomography). Approximately 95 percent of patients with a malignant nodule will have an abnormal FDG-PET (ie, sensitivity) and 78 percent of patients with a benign lung nodule will have a normal FDG-PET (ie, specificity) [43]. As a result, a negative FDG-PET will correctly exclude cancer in most cases (ie, high negative predictive value), but a positive FDG-PET will incorrectly identify infectious, inflammatory, or granulomatous nodules as malignant with considerable frequency (ie, poor positive predictive value) [44]. False-negative results can occur with tumors that have low metabolic activity (eg, bronchioloalveolar tumors, carcinoids, and some well-differentiated adenocarcinomas), small lesions (a critical mass of metabolically active malignant cells is required for detection by PET), and uncontrolled hyperglycemia because the uptake of FDG is retarded by the hyperglycemia [44]. An additional benefit of FDG-PET is the acquisition of staging data if the nodule is malignant. Nodule sampling — Consultation with a pulmonologist is warranted to decide whether and how the nodule should be sampled or excised if the clinical features, radiographic features, and quantitative models do not provide sufficient evidence that a lesion is either benign or malignant. Sampling of the nodule can be performed through the airway (fiberoptic bronchoscopy) or through the chest wall (percutaneous needle aspiration or biopsy). The choice of sampling procedure varies according to the size of the nodule, the location of the nodule, the availability of the procedure, and local expertise. Bronchoscopy — Fiberoptic bronchoscopy is a reasonable approach for the diagnostic evaluation of large, central nodules and masses; however, it is much less useful for small SPNs. When bronchoscopy is performed, specimens can be collected by washing or lavage of the airway supplying the lesion, or by direct sampling of the lesion with brushings or transbronchial biopsy. With all of these techniques, however, it is often difficult for the bronchoscopist to be certain about optimal positioning of the bronchoscope and placement of the brush or forceps within the lesion itself. As a result, the yield of bronchoscopy performed for diagnosis of a SPN is lower than other techniques, especially if the SPN is small. A systematic review of studies that examined the diagnosis of peripheral lung cancer using bronchoscopy with washing, brush, or biopsy, found an overall sensitivity of 69 percent [45]. The yield for diagnosing peripheral lesions smaller than 2 cm by brushing or biopsy was 33 percent, whereas the yield for lesions larger than 2 cm was 62 percent. The small amount of material provided by fiberoptic bronchoscopy is a limiting factor; thus, several approaches aimed at improving the diagnostic yield of bronchoscopy have been studied: - Fluoroscopy or endobronchial ultrasound can optimize localization of the lesion and increase the diagnostic yield of transbronchial biopsy (to greater than 70 percent) [46-49]. Unfortunately, these techniques require additional equipment and expertise and are not widely available. (See "Endobronchial ultrasound: Indications, advantages, and complications" and see "Endobronchial ultrasound: Technical aspects"). - Combining fluorescence in situ hybridization (FISH) and cytology may improve the sensitivity compared to cytology alone for both washings (61 versus 44 percent) and brushings (75 versus 51 percent); however, specificity is unaltered [50]. Validation of this technique is necessary before its routine use can be recommended. Percutaneous needle aspiration — Percutaneous needle aspiration (also called fine needle aspiration) of a SPN can be performed through the chest wall using either fluoroscopy or CT scanning to guide placement of the needle within the lesion. Its diagnostic yield is higher than fiberoptic bronchoscopy, since placement of the needle is more reliable. The overall diagnostic yield of percutaneous needle aspiration for suspected lung cancer is approximately 90 percent, and is higher when a cytologist attends the procedure than when a cytologist is not present to review the specimens [51,52]. Optimization of needle placement is particularly beneficial when evaluating a small SPN. An older study estimated that the diagnostic yield of percutaneous needle aspiration was 60 percent for SPNs smaller than 2 cm [53]. This was supported by a more recent study that reported a diagnostic yield of 64 percent for SPNs smaller than 3 cm [54]. In this study, the sensitivity and specificity of percutaneous needle aspiration for such nodules were 67 and 100 percent, respectively. Percutaneous needle aspiration obtains material for cytology, but does not biopsy a core of tissue. As a result, it is more helpful for confirming malignancy than establishing a specific benign diagnosis [33]. The former only requires the detection of malignant cells, while the latter usually requires evaluation of tissue architecture [33]. Percutaneous needle aspiration may be complicated by pneumothorax, which can be clinically significant in patients with coexisting emphysema. Bleeding is a less frequent complication. Percutaneous needle biopsy — An increasing number of centers are performing transthoracic needle biopsy to obtain a core of tissue, thereby improving the diagnostic yield of benign nodules. In this procedure, a core of tissue is obtained using a cutting needle. Up to 97 percent of patients with a malignant or benign lung nodule will obtain a definitive diagnosis using this procedure [55]. Surgical resection — Surgical excision of the nodule is indicated if the likelihood of malignancy is high (according to clinical and radiographic characteristics), an FDG-PET is positive, or a nodule has proven to be malignant by a sampling procedure. Excision can be performed by thoracotomy or thoracoscopy (also called video assisted thoracic surgery or VATS) if the lesion is located close enough to pleural surface to allow VATS resection [56,57]. The SPN is usually located by visual inspection. In some cases, however, the nodule is localized and marked preoperatively by placement of a hook wire (as used for localizing non-palpable breast lesions prior to biopsy) or by staining of the lesion with percutaneous injection of methylene blue [58]. For nodules sufficiently close to the pleura to be accessible to a video-assisted approach, VATS has essentially replaced thoracotomy, which has greater morbidity (eg, post-thoracotomy pain syndrome). VATS is very safe in experienced hands and can be performed as a day procedure at many centers. It has also replaced percutaneous needle aspiration or biopsy as a sampling technique in many medical centers, but recent studies have shown that approximately 50 percent of SPNs resected by VATs are benign lesions. That percentage of benign nodule resection is too high and dictates more rigorous preoperative evaluation. SUMMARY AND RECOMMENDATIONS A solitary pulmonary nodule (SPN) is a lesion that is both within and surrounded by pulmonary parenchyma. It is typically smaller than 3 or 4 cm. (See "Definition" above). The most common causes of a malignant SPN are primary lung cancer, carcinoid tumors, and lung metastases. Infectious granulomas and hamartomas are the most common causes of a benign SPN. (See "Differential diagnosis" above). Diagnostic testing cannot distinguish malignant and benign nodules unequivocally in all cases. An approach that leads to the excision of most malignant nodules is generally preferred, even if the cost is frequent resection of benign nodules. (See "General approach" above). During the diagnostic evaluation of a SPN, clinical features, radiographic features, and, occasionally, quantitative models are used to determine the probability that a nodule is malignant. (See "Diagnostic evaluation" above). Once the probability that a SPN is malignant has been formulated, initial management must be chosen. This choice should carefully consider patient preference. We propose the following approach to initial management, although we recognize that many reasonable alternative approaches exist (show figure 1): - We recommend surgical resection of a nodule that has a high probability of being malignant (Grade 1B). This recommendation reflects our belief that surgical resection is the only intervention that is potentially curative. (See "Initial management" above and see "Surgical resection" above). - We suggest following a nodule with serial CT scans if it has a low probability of being malignant, rather than surgically resecting the nodule immediately (Grade 2C). The frequency of CT scans varies according to the nodule's size and the patient's risk for lung cancer (show table 6). FDG-PET or sampling of the nodule are acceptable alternatives for patients who are uncomfortable with a strategy of observation. (See "Initial management" above and see "Serial CT scans" above). - We believe that a nodule that is 1 cm or larger with an intermediate probability of being malignant should be evaluated by FDG-PET, rather than followed with serial CT scans or surgically resected. We recommend that nodules that are positive on FDG-PET be excised (Grade 1B). Nodules that are negative on FDG-PET can be followed with serial CT scans. Sampling of the nodule should be performed if FDG-PET is unavailable. Sampling techniques include percutaneous needle aspiration, percutaneous needle biopsy, thoracoscopy, and bronchoscopy. (See "Initial management" above, see "Metabolic imaging" above, and see "Nodule sampling" above). - We suggest that a nodule smaller than 1 cm with an intermediate probability of malignancy be followed by serial CT scans, rather than excised, sampled, or evaluated by FDG-PET (Grade 2C). This suggestion reflects our recognition that the diagnostic yield of both sampling and FDG-PET is relatively poor for nodules smaller than 1 cm, and that there are risks associated with excision or sampling. (See "Initial management" above and see "Serial CT scans" above).