| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Present address: Department of Radiology, Christiana Hospital, 4755
Ogletown-Stanton Rd., Newark, DE 19713.
2 Department of Radiology, The Children's Hospital of Philadelphia,
Philadelphia, PA.
3 Present address: Department of Pediatrics, Vanderbilt Children's Hospital,
Nashville, TN.
Received October 1, 2004;
accepted after revision November 23, 2004.
Address correspondence to G. E. Wilde
(gregwilde{at}comcast.net).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Symptomatic PTLD in children manifests earliest in lung recipients and can involve any organ system. However, PTLD in the thorax is most common after lung transplantation, and PTLD in the abdomen most commonly follows kidney transplantation.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The linkage of the PTLD process to the immunosuppressed state is further suggested by the occurrence of 80% of cases within the first year of organ transplantation during peak immunosuppression [2]. The time from transplantation to diagnosis of PTLD in the pediatric population has ranged from 6 weeks to 7 years [2]. Although any organ system may be involved at presentation, reports from the literature have found the abdomen to be involved in 60-70%; the thorax, in 45-65%; the head and neck, in 20-30%; and the brain, in 1-10% [2].
Because earlier diagnosis of PTLD can improve the response to therapy, most transplantation centers rely on radiologic screening at preset time points after transplantation to aid in the detection of asymptomatic disease. To this end, recognition of the most salient imaging manifestations of PTLD has been a goal of several previous reports in the literature [4-7]. Some of these researchers have also attempted to correlate the anatomic location of presentation and time to onset of PTLD with the type of allograft received.
Donnelly et al. [4] first reported a linkage between location of PTLD and allograft type when they found that PTLD tended to occur in the anatomic region of the transplant among heart, liver, and kidney recipients. Lim et al. [5] found that lung recipients manifested PTLD earlier than heart recipients and that disease was more common in the thorax among lung recipients. Pickhardt and Siegel [6] found that abdominal PTLD was more common in liver and heart recipients than lung and kidney recipients. However, these studies either have failed to incorporate all major allograft types in their study population or have failed to examine all anatomic regions. Whether the conclusions of these studies are broadly applicable to all transplant types has never been validated in the pediatrics literature.
The objective of this retrospective study was to correlate the anatomic location, extent, and time to onset of PTLD in a study population of pediatric heart, lung, liver, and kidney allograft recipients. These correlations would be helpful for adjusting CT surveillance protocols for asymptomatic patients according to allograft type; they also would be useful to pediatric radiologists in the interpretation of surveillance imaging of asymptomatic patients.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Thirty-one of these cases were identified as having pathologically proven PTLD. PTLD was pathologically proven if tissue obtained by open surgical biopsy, percutaneous biopsy, endoscopic biopsy, or autopsy revealed a diagnosis of PTLD based on the consensus classification system. The consensus classification system categorizes a histopathologic diagnosis as benign polyclonal proliferation, polymorphic proliferation, or monomorphic proliferation in ascending order of severity [8]. Of those with pathologically proven PTLD, patients were excluded if they survived for less than 1 month after transplantation, if there were no imaging findings, or if they were older than 20 years old at the time of transplantation. One of the 31 patients identified was excluded because of an absence of imaging findings. Although the presence of symptoms was not a criterion for inclusion, all 30 remaining patients were symptomatic at presentation: six lung or heart-lung recipients, nine heart recipients, eight liver recipients, six kidney recipients, and one bone marrow recipient (Table 1).
|
The lung transplant and heart-lung transplant recipients were combined into a single group. This combination was done, in accordance with previous protocols, because heart-lung recipients require the same level of immunosuppression used for lung recipients [5]. Although the imaging findings for the bone marrow transplant recipient are presented, this case was excluded from all statistical analyses. Two authors retrospectively reviewed the medical records to determine the type of transplant, time to diagnosis after transplantation, signs and symptoms at presentation, involved sites, and histopathologic subtypes. When available, the histopathologic subtype of PTLD was recorded.
Images were reviewed by one author. All imaging studies for each patient, including both conventional radiographs and cross-sectional images, were evaluated together. In most cases, however, findings are presented from the cross-sectional studies. Thoracic images were evaluated for the presence of parenchymal nodules or consolidation, pleural disease, and mediastinal adenopathy or masses. Abdominal images were reviewed for the presence of visceral, nodal, and extranodal disease. We evaluated all head and neck images for the presence of oropharyngeal lymphatic disease or cervical adenopathy. All central nervous system images were reviewed for abnormalities of the parenchyma, soft tissues, or bones.
An imaging finding was attributed to PTLD when there was a corresponding direct pathologic correlation, when there was a direct pathologic correlation in the same organ or anatomic region, or when lesions decreased in size on subsequent imaging studies in response to treatment, as defined by prior protocols [5]. Whenever possible, additional pathologic data were included if tissue had been obtained from multiple potentially involved sites with upper or lower endoscopy (or both), core needle biopsy with CT guidance, or excisional biopsy.
Regions of presentation were divided anatomically into central nervous system, head and neck, chest, and abdomen and pelvis. Statistical analysis was undertaken to determine whether there were significant differences between transplant types in distribution of disease and time to diagnosis after transplantation. Nonparametric methods for testing differences between two groups were used. To make two populations, we compared each transplant type with all other transplant types combined in a single group. For comparing distributions of disease, 2 x 2 tables were formed for each transplant type versus anatomic region. The frequency of involvement of a specific anatomic region was compared with the frequency of all disease involvement outside that anatomic region (Table 2). The 2 x 2 categoric data were analyzed using the Fisher's exact test. For comparing time to diagnosis of PTLD, the Mann-Whitney U test for continuous variables was used. For both tests, p values less than 0.05 were taken to indicate significance.
|
This study was approved by the hospital's institutional review board. The need for individual informed consent was waived.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Symptomatic PTLD tended to present earlier in heart-lung (median, 6.4 months; range, 3-34 months) than in heart (median, 22.3 months; range, 6-69 months), kidney (median, 27.9 months; range, 6-62 months), or liver (median, 22.0 months; range, 4-108 months) recipients. A Mann-Whitney U test comparing the heart-lung group and the lung group with all other groups yielded a p value of 0.036 (Fig. 1).
|
|
Abdominal PTLD occurred in 17 (57%) of the 30 cases. The most frequent findings were bowel wall thickening (n = 12) and abdominal lymphadenopathy (n = 9; patients 6, 8-10, 15, 16, 25, 26, and 28 in Table 1). Bowel wall thickening (Fig. 3) was one of several manifestations of the 12 cases of small bowel involvement on both CT and upper gastrointestinal series; other presentations included perforation, obstruction (Fig. 3), and nodular polypoid lesions in both the duodenum and distal ileum. Nine patients presented with retroperitoneal or mesenteric lymphadenopathy. Most of these patients had discrete lymph node involvement, but in five patients, extensive adenopathy coalesced to form large retroperitoneal masses (Figs. 4A and 4B).
|
|
|
Abdominal PTLD occurred more commonly in the kidney recipients (n = 5, 83%) than the heart (n = 5, 56%), liver (n = 4, 50%), and lung or heart-lung (n = 2, 33%) recipients. Of the kidney recipients, only two (33%) of six had other regions involved. Comparison of the abdominal findings for kidney recipients with those for recipients of all other transplants did not yield a statistically significant difference between the groups (p = 0.136).
Nine (30%) of the 30 cases presented with disease of the head and neck. Patients in our study most commonly presented with diffuse enlargement of the tonsils (n = 3; patients 14, 18, and 29 in Table 1) or cervical adenopathy (n = 4; patients 1, 7, 17, and 22 in Table 1). Figure 5 shows the CT findings of a patient with coalescence of cervical and supraclavicular adenopathy. Two patients in our study presented with a nonspecific fever and recurrent sinusitis and were subsequently found to have PTLD by nasal polyp biopsy. In one case, the orbit was involved; CT showed a well-defined area of focal soft-tissue swelling within the right infraorbital preseptal soft tissues. Head and neck disease occurred with similar frequency among the solid organ transplant groups, and none of the comparisons met criteria for statistical significance accordingly.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our data are in partial agreement with those from these prior studies. Our study findings show important trends in anatomic manifestations of disease among pediatric heart-lung recipients. There was a clear tendency for thoracic PTLD, particularly parenchymal disease, to develop preferentially after lung transplantation. Our finding of 100% of lung recipients with PTLD to have intrathoracic presentation is comparable to the 89% reported by Lim et al. [5], 69% reported by Siegel et al. [7], and 100% reported by Montone et al. [9].
We also confirm a trend among lung transplant recipients in the relative frequency of mediastinal versus parenchymal disease. In lung transplant recipients, parenchymal disease has been four times more common than mediastinal disease [10]. Our study supports those findings in that all six lung recipients presented with parenchymal disease, and only one had concurrent mediastinal disease. In contrast, among nonlung recipients, the parenchymal and mediastinal regions have been involved with equal frequency [10]. In our study, mediastinal disease was slightly more common among nonlung recipients. Eleven of 12 patients had mediastinal or axillary node involvement, whereas seven of 12 had parenchymal disease.
There was also a trend for abdominal PTLD to develop more frequently after kidney transplantation than after transplantation of other organs. This is in contrast to the results of Pickhardt and Siegel [6] who found that abdominal PTLD is more common among heart (100%) and liver (88%) recipients than lung (61%) and kidney (53%) recipients. In our study, liver and heart transplants appeared to be the exception to the predilection for PTLD to involve the allograft and adjacent regions. These recipients presented with near equal frequency in each anatomic region. Among the heart transplants, none had allograft involvement, 78% had intrathoracic PTLD, and 56% had abdominal PTLD. Among the liver transplants, none had CT evidence of allograft involvement, although two of eight patients did show signs of liver failure, 63% had intrathoracic PTLD, and 50% had abdominal PTLD.
The most important finding in our study involves the timing of symptomatic PTLD after transplantation. The onset of symptomatic PTLD was sooner in lung transplant recipients than in other groups in our study (p < 0.05). This finding adds to those of Lim et al. [5] and Armitage et al. [11]; our study confirms the onset of symptomatic disease in heart-lung recipients is earlier even when liver and kidney transplants are included in the analysis.
Although our study was able to suggest a number of trends in anatomic distribution and make a definitive conclusion regarding time to onset, the rarity of PTLD prevented recruitment of an adequate number of subjects for achieving the statistical power necessary to make statistically significant conclusions about disease distribution. Perhaps a study across multiple institutions would achieve the
necessary power to evaluate data with more robust parametric methods.
Because a strong risk factor for PTLD involves the level of immunosuppression imposed on a transplant recipient, it lends to reason that the type of immunosuppressive regimen used for each of the patients in our study could correlate well with specific imaging findings for certain transplant types, extent of disease at presentation, and time to disease onset. Prognosis may also correlate with specific imaging findings, extent, or location of disease; information about prognosis could be useful for guiding treatment decisions. These issues could be evaluated in a future study.
Screening surveillance protocols after transplantation vary widely across institutions. A prospective study could be undertaken to compare screening protocols on a population of transplant recipients divided into groups according to transplant type and immunosuppression type.
Based on the findings presented in this study, one might hypothesize that lung transplant recipients need earlier and more frequent screening of the chest with occasional screening of the abdomen and vice versa for recipients of kidney allograft. The cost effectiveness of these various screening regimens could be examined, and correlation of initial imaging findings with prognosis and survival could also be made.
In summary, PTLD seems to be most common when higher levels of immunosuppression are necessary and in patients who are EBV-seronegative before transplantation. As a result, pediatric recipients of organs that require high levels of immunosuppression are the most susceptible population. In particular, when recipients of lung allograft are compared with recipients of all other major allograft types, lung recipients have an increased risk for involvement of the allograft itself or the mediastinum and for an earlier onset of symptomatic disease after transplantation. Also, kidney recipients seem to have an increased risk for involvement of the abdomen by PTLD. However, more research is needed to confirm the data presented in this study and determine the most likely distributions of PTLD for liver and heart recipients. With increasing recognition of the distribution and timing of PTLD according to allograft type and perhaps to level of immunosuppression, more cases will be diagnosed promptly, perhaps leading to more frequent successes in the treatment of patients with PTLD.
Acknowledgments
We thank Rochelle Tractenberg, Department of Biostatistics, Georgetown
University, for her assistance with statistical methodology.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
