There have been only a few publications on the subject of primary neoplasms of the diaphragm, reflecting the rarity of these tumors. Grancher¹⁴ is commonly credited with providing the first description in 1868 of a primary diaphragmatic tumor, a benign fibroma, unsuspected until discovered at autopsy. Other authors give primacy to Clark⁹ for a report in 1886 of a primary lipoma of the diaphragm. Isolated case reports have predominated subsequently; these have been summarized in four major reports. Three of these, which review adult cases, have been published by Nicholson and Whitehead,²³ Wiener and Chou,³³ and most recently Olafsson,²⁴ with the last report finding a total of only 71 case reports within the world literature until that time. Since 1971, there have been approximately 40 additional reports, which, together with the cases in the review articles, serve as the basis for this chapter. In 2006, Cada⁵ reviewed five pediatric patients seen at their institution and 36 other patients from prior reports. Approximately 28 case reports have been published since 1995; this increasing rate of publication may possibly reflect the more common use of imaging and detection of asymptomatic lesions.

A review of the case reports available in the literature suggests that diaphragmatic tumors are not associated with any characteristic symptoms. Approximately 50% are asymptomatic, discovered either incidentally on radiographs or at surgical exploration performed for unrelated reasons. If any symptom is characteristic, it would be lower chest discomfort or heaviness, possibly with referred pain to the shoulder. Larger masses may give rise to compression of adjacent structures—such as the lung—causing cough, dyspnea, hemoptysis (rarely), or lower extremity edema if venous return is compromised. Other physical findings appear uncommon, although some authors have noted masses palpable in the upper abdomen or bulging through rib interspaces.

Evaluation prior to surgery is almost exclusively radiographic. Older reports, such as that of Ackerman¹ in 1942, suggested fluoroscopy in conjunction with the performance of diagnostic pneumoperitoneum and pneumothorax. Subsequent reports suggested the use of angiography. These measures are rarely used today, having been replaced by ultrasound (US), computed tomography (CT), and magnetic resonance imaging (MRI), as discussed by Yeh,³⁴ Brink,⁵ and Kanematsu,¹⁸ respectively.

Given the rarity of diaphragmatic lesions, it is reasonable to suggest that after a lesion is detected in the vicinity of the diaphragm, initial evaluation should be directed toward the possible presence of more common benign or malignant lesions arising in proximity to the diaphragm but not from it. In addition, the presence of normal diaphragmatic anatomic variants that may be confused with neoplasms should also be considered. In fact, it is likely that the diaphragm will not even be considered as the source of a lesion. Cada⁶ noted that only 6 out of 18 pediatric patients with primary diaphragmatic lesions had the diaphragm identified as the source after radiologic evaluation and before surgery.

One must eliminate other possibilities such as nonneoplastic disease or masses arising from structures near the diaphragm. Nonneoplastic processes include infections (e.g., tuberculosis), hematomas, congenital abnormalities, or hernias. Organs in the region of the diaphragm may be involved by processes that may be confused with primary diaphragmatic processes. In particular, infections or primary or secondary neoplastic diseases of the liver, lung, pericardium, thymus (in particular, thymolipoma), stomach, or spleen should be kept in mind. In addition, neoplasms may invade or metastasize to the diaphragm (e.g., malignant pleural or peritoneal processes).

Usually, the initial test suggesting the presence of a lower thoracic cavity density is a chest radiograph. Decubitus films may eliminate the presence of a free-flowing effusion. Further evaluation will usually consist of some combination of CT, MRI, and/or US. Structural abnormalities of the diaphragm—such as lobulations, localized eventrations, slips, and hypertrophic crus—may simulate neoplasms. However, these abnormalities can usually be distinguished from neoplastic processes by their appearance. For example, Ferguson¹⁰ and Yeh³³ discuss the fact that lobulations are usually multiple and that eventrations generally conform to or retain the overall shape of the diaphragm. Distinguishing a mass apparently arising from the diaphragm from one originating from within the lung parenchyma may be facilitated by reconstructions of computed tomographic
images provided that a low-attenuation plane can be seen between the mass and the diaphragm. The radiology literature indicates that imaging may support the localization of a mass within the lung by suggesting irregular margins with the nearby lung, the presence of acute angles between the diaphragm and the mass, and focal volume loss in the lung, with pulmonary vessels and bronchi appearing to curve into the lesion. A pleural origin for a mass is suggested by the presence of obtuse angles between the lesion and the diaphragm.

Imaging of herniated hollow viscera through the diaphragm may be enhanced by the use of orally or intravenously administered contrast. US or MRI of the liver may be helpful in localizing a mass to the liver parenchyma. The choice from among the wide range of currently available diagnostic imaging modalities will depend largely on the suspected diagnosis, but to some extent extensive radiographic investigations may be replaced by video-assisted thoracic surgical techniques (VATS), and/or laparoscopy, as these offer the possibility of both diagnosis and therapy. There is no defined role for more recent imaging modalities, such as positron emission tomography (PET).

The most commonly reported benign neoplasm of the diaphragm is the lipoma. Other reported benign diaphragmatic lesions include fibromas and congenital cysts. Case reports of chondromas, angiomas, lymphangiomas, hemangioendotheliomas, rhabdomyofibromas, and neurofibromas have also been published.

As noted above, the first reported case of a diaphragmatic lipoma dates to Clark⁹ in 1886, with approximately 25 subsequently reported cases. Over half of these cases were previously unsuspected findings at autopsy. The symptoms attributed to these lesions in the remaining patients have been largely those of vague lower chest wall discomfort. There has been one reported case in the pediatric population,⁶ but given that the majority of primary diaphragmatic neoplasms come to light at either autopsy or as asymptomatic radiographic findings, it is possible that they are present but undiagnosed in the pediatric population.