CORRESPONDENCE.

(Letters to the Editor are welcomed. They may report new clinical or laboratory observations and new developments in medical care or may contain comments on recent contents of the Journal. They will be published, if found suitable, as space permits. Like other material submitted for publication, letters must be typewritten, double-spaced, and submitted in duplicate. They must not exceed two typewritten pages in length. No more than five references and one figure or table may be used. See "Information for Authors" for format of references, tables, and figures. Editing, possible abridgment, and acceptance remain the prerogative of the Editors.)

Intravascular Lymphoma Presenting With Clinical Features of Cholecystitis 

To the Editor: Intravascular lymphoma, also known as malignant angioendotheliosis or angiotropic large-cell lymphoma, is a rare lymphoma characterized by a proliferation of malignant lymphoid cells within vascular lumen with secondary intravascular thrombi resulting in obliteration of the involved vessels. The disease was first described by Pfeger and Tappeiner in 1959 as "angioen-dotheliomatosis proliferans systemisata" (Zur kenntnis der systemisierten Endotheliomatose der cutanen Blutgefasse) and predominantly involves vessels in the skin and central nervous system, although blood vessels in other organs may also be affected. The disease may produce a variety of clinical pictures, but most often patients present with central nervous system and cutaneous abnormalities. [1] Involvement of the gallbladder is a rare event. We report a case of intravascular lymphoma involving the gallbladder.

CASE REPORT

A 64-year-old woman presented to a hospital in Mexico City with a 3-month history of fever, night sweats, weight loss, and fatigue. Three days before admission, she had jaundice, acholia, choluria, and disorientation. Physical examination showed conjunctival jaundice, hepatosplenomegaly, and edema of both legs. Neurologic examination showed fluctuating mental status, aggressive behavior, and slurred speech. There were nonfocal signs. Laboratory evaluation showed anemia, thrombocytopenia, leukocytosis, and transaminasemia. Cerebrospinal fluid examination revealed elevated proteins with normal glucose and cell counts. A CT scan demonstrated splenomegaly, and ultrasound showed thickening of the gallbladder wall and biliary sludge. The intrahepatic and extrahepatic bile ducts were normal. Open cholecystectomy with transcystic cholangiography showed an edematous gallbladder with a normal extrahepatic biliary tree. There were no calculi. Histologically, the gallbladder showed vascular dilatation of the lamina propr ia, with numerous large intraluminal cells with little cytoplasm and hyperchromatic, pleomorphic nuclei with few mitoses (Figures 1 and 2). The neoplastic cells were positive for CD45 (LCA), CD20 (L-26) (Dako), CD79a (Cell Marquee) and bcl-2 (Biogenex) and showed kappa (Dako) light chain restriction. No staining of tumor cells was noted with keratin AE1-3, EMA, and LMP-1 (Dako) (Figure 3). After surgery, the patient's clinical status deteriorated rapidly, and she died after 5 days of hospitalization. Postmortem examination was not performed.

DISCUSSION

Intravascular lymphoma is a rare disorder in which there is a proliferation of malignant lymphoid cells within the lumina of small and medium-sized vessels. The disease predominantly affects CNS and skin, but other organs, such as the gallbladder, may be affected, as shown in this case. The neoplastic cells were originally believed to be of endothelial origin, but inmunohistochemical and gene rearrangement studies have demonstrated the lymphoid nature of the neoplastic cells. Although most of the reported cases are B-cell lymphomas, T-cell and even histiocytic lineage have also been reported. [2] The disease follows a subacute course, usually culminating in death within several months to years after the onset of symptoms. Clinical features often include behavioral or personality changes progressing to dementia, associated with focal neurologic signs of a fluctuating nature. Other manifestations are much less constant and reflect variable anatomic sites of involvement. In our case, the main manifestations wer e abdominal pain, probably due to the intravascular lymphoma present in the gallbladder, and neurologic signs. Although the CNS and the skin are the most commonly affected organs, autopsy studies have shown that intravascular lymphoma may involve the small vessels of various organs. [1,3,4]

In summary, intravascular lymphoma is a rare form of lymphoma, presenting with a diverse and confusing clinical picture. To date, most cases have been diagnosed during the postmortem review, and rarely the gallbladder has been affected. This may be due to the few autopsy cases studied. Complete remission and sometimes long-term disease-free survival may be obtained with standard combination chemotherapy designed for high-grade lymphoma. [4,5] The overall mortality rate is above 80%, and the average survival time is 13 months.

Fabiola Flores-Vazquez, MD

Department of Internal Medicine

Beatriz de Leon-Bojorge, MD

Carlos Ortiz-Hidalgo, MD

Department of Pathology

Marino Capurso, MD

Department of Surgery

American British Cowdray Hospital

Sur 136 Esq Observatorio

Mexico City DF 01120

Mexico

References

(1.) Demirer T, Dail H, Aboulafia D: Four varied cases of intravascular lymphomatosis and a literature review. Cancer 1994; 73:1738-1745

(2.) O'Grady JT, Shahidullah H, Doherty VR, et al: Intravascular histiocytosis. Histopathology 1994; 24:265-268

(3.) Theaker J, Gatter K, Esiri M, et al: Neoplastic angioendotheliosis - further evidence supporting a lymphoid origin. Histopathology 1986; 10:1261-1270

(4.) Sheibani K, Battifota H, Winberg C, et al: Further evidence that "malignant angioendotheliomatosis" is an angiotropic large-cell lymphoma. N Engl J Med 1986; 314:943-948

(5.) Fulling K, Gersell D: Neoplastic angioendotheliomatosis. Cancer 1983; 51:1107-1118

Striae Gravidarum as a Predictor of Vaginal Lacerations at Delivery

To the Editor: I have many concerns about the recent article by Wahman et al. [1] It is a study attempting to determine whether striae gravidarum are associated with a tendency to perineal lacerations--an interesting hypothesis. Unfortunately, the analysis is bedeviled by a series of methodological problems. Fundamental is the failure to recognize that episiotomy itself is a planned second degree tear. The statement in the abstract that episiotomy itself prevents spontaneous lacerations is therefore misleading. While perhaps 50% of all births conducted without episiotomy will require some suturing, every episiotomy will require suturing. The issue is whether the woman is better off with or without episiotomy, and the authors' data fail to address that issue. Our own data and those of others fail to show a benefit to routine use. [2-4] An association does not necessarily imply the need for an action to correct an association. Thus, to suggest that the presence of striae might be an indication for episiotomy i s unsupported.

Specifically, in order to understand the issue more clearly, a correction of Table 2 is needed. It is misleading to present lacerations without counting episiotomy as a deliberate second degree tear. Thus, if this Table were reworked to take this into account, the apparent slight association between striae and lacerations would disappear. Similarly, Table 3, if reworked to use need for suturing as the important outcome, would see the category "lacerations requiring repair" vanish into meaninglessness. Finally, Table 4 is not useful to make the point that episiotomy "prevents" lacerations, as it again does not present the most important outcome.

In summary, the statistical associations, which are weak at best, make no clinical sense and risk encouraging clinicians to consider a "new" indication for episiotomy when none exists.

Michael C. Klein, MD, CCFP, FCFP, FCPS

Department of Family Practice and Pediatrics

University of British Columbia 

Children's and Women's Health Centre of British Columbia

Maternity and Newborn Care Committee College of Family Physicians of Canada

4500 Oak St, Room F412B

Vancouver, British Columbia V8H 3N1 Canada

References

(1.) Wahman AJ, Finan MA, Emerson SC: Striae gravidarum as a predictor of vaginal lacerations at delivery. South Med J 2000; 93:873-876

(2.) Klein MC, Gauthier RC, Jorgenson SH, et al: Does episiotomy prevent perineal trauma and pelvic floor relaxation? On-line Journal of Current Clinical Trials. American Association for the Advancement of Science. 1992: 1;doc 10

(3.) Klein MC, Cauthier RC, Robbins J, et al: Relationship of episiotomy to perineal trauma and morbidity, sexual dysfunction, and pelvic floor relaxation. Am J Obstet Gynecol 1994; 171:591-596

(4.) Klein MC, Janssen PA, MacWilliams L, et al: Determinants of vaginal/perineal integrity and pelvic floor functioning in childbirth. Am J Obstet Gynecol 1997; 176:403-410

Reply

To the Editor: Dr. Klein's comments are pertinent and appreciated. Dr. Klein and his associates have extensive support in the literature for their opinion; their references are impressive. I do, however, respectfully disagree that "episiotomy itself is a planned second degree tear." Episiotomy is, in fact, a surgical incision designed to control the size of the delivery outlet at the level of the perineum/pelvic floor. Dr. Klein also misses the most important point of our manuscript: We noted an interesting association between striae gravidarum and vaginal lacerations at the time of vaginal delivery. To our knowledge, this curious association between striae gravidarum and the elasticity of the perineum has never been previously reported in the medical literature. This association held up through a rigid and thorough multivariate logistic regression analysis. The univariant analysis presented in Tables 2 and 4 were simply provided to demonstrate a small portion of our data. They were not meant to be interpret ed as a stand-alone analysis. All of these data were considered in our multivariate logistic regression analysis, which is the primary support of our conclusions. In our opinion, to rework the univariant analysis tables as Dr. Klein suggested would be a misrepresentation of our data. This analysis included controlling for dozens of variables. A multivariate logistic regression analysis cannot be reproduced easily in a Table. However, the association between striae gravidarum and perineal lacerations held up throughout our multivariate logistic regression analysis with a Pvalue of .00002.

Dr. Klein's comments with regard to reworking Table 2 are not pertinent, since an episiotomy is not, in fact, a deliberate second degree tear, but is actually a surgical incision that is performed under controlled conditions. Additionally, lacerations that do not require suturing are important to the patient and might not only increase postdelivery recovery, but might be associated with longterm pelvic floor relaxation. They also might be a result of poor skin elasticity, which we have shown to be associated with striae gravidarum. I maintain that this does, in fact, "make clinical sense."

I strongly differ with Dr. Klein's last statement that our statistical associations are "weak at best." A highly significant Pvalue that held up throughout multivariate logistic regression analysis is considered a strong association. I would urge Dr. Klein, as well as the other interested readers of this Journal, to attempt to reproduce our results in order to confirm the association between striae gravidarum and vaginal lacerations at delivery. Our concluding statement that "further research is necessary to determine whether there is a true correlation and to analyze other factors that may act as predictors" might be the most salient point. The readers of this Journal and academicians might be of some assistance to patients by further trying to confirm or refute our findings.

Michael A Finan, MD

Department of Obstetrics and Gynecology 

Division of Gynecologic Oncology

Ochsner Clinic and Alton Ochsner Medical Foundation

1514 Jefferson Highway

New Orleans, LA 70121-2483