We describe the first recognized serious complication of a measles vaccine virus (fatal giant-cell pneumonitis) in a young male vaccine recipient with AIDS.

Case Report

Top Case Report Results Discussion Author & Article Info References

A 21-year-old man with hemophilia A, AIDS, and an undetectable CD4 cell count presented on 31 August 1993 with progressive cough, dyspnea, and fever. Infection with HIV had been documented 3 to 4 years previously, and the patient had had Pneumocystis carinii pneumonia in October 1992. He had received a booster immunization with measles-mumps-rubella (MMR) vaccine in September 1992, as required for college entry.

On physical examination at hospital admission, the patient appeared ill and was febrile. Abnormal physical findings were limited to oral thrush and basilar pulmonary rales. Chest radiography revealed several bilateral, poorly marginated parenchymal nodules.

After nondiagnostic bronchoalveolar lavage and transbronchial biopsies, thorascopic lung biopsy was performed on 6 October 1993. Histopathologic examination revealed numerous multinucleated giant cells, some of which contained intracytoplasmic and intranuclear inclusions suggestive of measles giant-cell pneumonia (Figure 1). Transmission electron microscopy showed particles that were morphologically consistent with measles virus. On 22 October 1993, tissue culture cells inoculated with lung tissue displayed cytopathic changes that were characteristic of measles virus infection and were confirmed by immunofluorescence microscopy with a measles-specific monoclonal antibody. The patient was negative for serum antimeasles IgM titers and positive for IgG antibodies (1:5), but the latter test was performed after intravenous immune globulin had been administered.

View larger version (123K): [in this window] [in a new window]   Figure 1. High-power view of lung biopsy tissue showing nodular areas of acute and chronic inflammation with regions of necrosis and fibrosis. Multinucleated giant cells are present, some of which contain both nuclear inclusions (thin arrows) and cytoplasmic inclusions (broad arrow). (Hematoxylin-eosin; original magnification, x500.).

On the basis of the diagnosis of measles giant-cell pneumonitis, the patient was treated with intravenous immune globulin and oral ribavirin. Despite this treatment, the patient's condition progressively deteriorated; he died on 17 December 1993.

Results

Top Case Report Results Discussion Author & Article Info References

Isolation of measles virus from lung tissue with histologically established giant-cell pneumonitis clearly defined the etiologic basis of this patient's disease. However, the possibility that this measles virus isolate was anything other than wild-type was not considered until a delimited sequence analysis of its genome revealed a "signature" pattern previously seen only in vaccine viruses. The entire genome of the isolate and that of the currently used vaccine strain, Moraten (Attenuvax, Merck, Sharpe, & Dohme, West Point, Pennsylvania), were subsequently sequenced. Comparison of these two genomes, which were 15 894 nucleotides long, showed that they were almost identical. Only two nucleotide differences were found (Figure 2). An A at nucleotide 8905 of the Moraten genome had changed to a G in the patient's virus, causing a serineglycine substitution at amino acid position 546 of the hemagglutinin protein. The other difference lay within the polymerase (L) gene at position 13 369, where a G in the vaccine virus was changed to an A; this caused an aspartateasparagine change in the patient's virus isolate. Analysis of the measles virus-specified RNA extracted directly from the patient's diseased lung tissue revealed the same two nucleotide differences.

View larger version (11K): [in this window] [in a new window]   Figure 2. Nucleic acid and amino acid differences between Moraten vaccine virus and the patient's measles virus isolate. The complete genomic sequences of both viruses were determined. Total RNA was extracted from the patient's lung biopsy material, from tissue culture cells in which the patient's virus isolate was propagated, and from the commercially available Moraten vaccine virus (Attenuvax, Merck, Sharpe, & Dohme, West Point, Pennsylvania) by using TRIzol LS reagent (Life Technologies, Grand Island, New York) according to the manufacturer's protocol. The RNA was reverse transcribed to complementary DNA by using random hexamers and Murine Moloney Leukemia Virus Reverse Transcriptase (Perkin-Elmer Cetus RT-PCR kit reagents, Perkin-Elmer Cetus, Branchburg, New Jersey). The complementary DNA was then amplified by polymerase chain reaction (PCR) to create overlapping fragments that spanned the entire 15 894 nucleotide-long measles genome by using measles virus-specific oligodeoxynucleotide primer pairs whose design was based on the known Edmonston measles virus sequence [8]. The PCR products were directly sequenced, without cloning, to obtain a consensus sequence by the dideoxy terminator cycle-sequencing method (ABI PRISM 377 sequencer and the ABI PRISM sequencing kit; Perkin-Elmer Cetus, Foster City, California). Both strands of the DNA were sequenced to eliminate possible ambiguities. The only two differences, out of the total 15 894 bases determined for each viral genome, are shown. The open-reading frames (boxes) coding for the N (nucleocapsid), P (phosphoprotein polymerase cofactor), M (matrix), F (fusion), H (hemagglutinen), and L (polymerase) proteins and the leader, trailer, and intercistronic regions (lines) are delineated in the measles virus genome (negative sense RNA) shown. All distances are proportional to their actual size within the genome. Arrows indicate the position of the nucleotide differences between the genome of the patient's measles virus and that of the Moraten vaccine virus. The nucleotide positions are given in reference to the genomic position (out of 15 894 bases total). The amino acid positions are based on the coding potential of the H (618 aa) and L (2183 aa) open-reading frames.

Discussion

Top Case Report Results Discussion Author & Article Info References

Universal immunization with live, attenuated measles virus vaccine has proven to be a remarkably safe and effective means for preventing measles infection and disease, even in immunocompromised recipients. Only one instance of Moraten vaccine-associated death, which occurred in a 6-month-old child with congenital combined immunodeficiency, has been reported previously [9].

Although the vaccine origin of our patient's giant-cell pneumonitis was established by genomic sequence analysis, its source remains perplexing. Previous studies of parenterally administered measles vaccine showed no pulmonary shedding or transmission to vaccine-naive contacts [10, 11], and no intimate, sustained contact between the patient and a recent vaccinee was identified. The only obvious remaining source was the MMR booster immunization received a year before apparent disease developed. Speculation that wild-type and vaccine measles viruses may establish persistent infection of lymphoid cells has been long-standing, but supporting evidence is sparse. A study of HIV-infected children who received measles vaccine virus did not show persistent infection of peripheral blood mononuclear cells [12]. However, examination of these cells in vaccinated children with autoimmune hepatitis demonstrated that vaccine virus persisted for as long as 7 years [13].

This case was reported by the Centers for Disease Control and Prevention (CDC) in July 1996 [14]; as a result, the ACIP is reevaluating its measles immunization recommendations. The CDC has also stated that "in the interim, it may be prudent to withhold MMR or other measles-containing vaccines from HIV-infected persons with evidence of severe immunosuppression" [14]. More recently, the Committee on Infectious Diseases of the American Academy of Pediatrics has recommended that "severely immunocompromised patients with HIV infection ... should not receive measles vaccine" [15]. The overwhelming success of measles immunization programs must be strongly supported. This case merely underscores the need for alternative ways to achieve immunoprophylaxis in patients with significant immunologic dysfunction.

Drs. Walpita, Lerch, Sidhu, and Udem: Wyeth-Lederle Vaccines & Pediatrics, 401 North Middletown Road, Pearl River, NY 10965.

Dr. Masurekar: Division of Infectious Diseases, New Jersey Medical School, 185 South Orange Avenue, Newark, NJ 07103.