Bone Marrow Transplantation (2000) 26, 587–589  2000 Macmillan Publishers Ltd All rights reserved 0268–3369/00 $15.00 www.nature.com/bmt Correspondence Early viral complications after autologous CD34-selected peripheral blood stem cell transplantation We considered of great clinical significance a recent report from Holmberg et al 1 on the increased incidence of CMV disease after autologous CD34-selected peripheral blood stem cell (PBSC) transplantation. We reported previously an increased incidence of viral complications after autolog- ous CD34-selected PBSC transplantation in 19 consecutive patients. 2 We recently analyzed a cohort of 40 CMV- seropositive patients affected by hematological malig- nancies submitted to CD34-selected PBSC. CD34 selection was carried using the Ceprate SC (Cellpro, Bothell, WA, USA) in 25 patients and CliniMACS (Miltenyi Biotec, Bergisch-Gladbach, Germany) in 15 patients thereafter, both according to the manufacturer’s specifications and we retrospectively evaluated the incidence of CMV disease and other viral complications within the first 100 days after transplantation. Results were compared to a cohort of 136 CMV-seropositive patients submitted to unfractionated PBSC and treated at the same institution over a period of 5 years. Overall 31 patients (77.5%) submitted to CD34- selected PBSC transplantation had weekly CMV screening studies performed including the CMV pp65 antigenemia assay (CMV Brite; Biotest Diagnostic Corp, Denville, NJ, USA) and viral culture monitoring from stools, urine and throat were performed weekly in 30 patients. CMV infec- tion was defined as either the evidence of any level of quan- titative pp65 antigenemia and/or a positive culture. CMV disease was defined as a conventional culture of bronchoal- veolar lavage fluid, or gastric/duodenal biopsy in associ- ation with symptoms. 3 Other viral infections were defined as the evidence of positive culture from urine, stools or throat in association with symptoms. Of the 40 patients, 39 were transplanted for hematological malignancies (97.5%) including NHL, MM and HD. The median number of CD34 + × 10 6 /kg infused was 4.2 (0.86–16). The character- istics of the patients are shown in Table 1. Of the 136 patients transplanted with unfractionated PBSC, 134 were affected by hematological malignancies including NHL, MM, HD and AML. The CMV screening study was carried out in 105/136 patients (77%). In this group the median number of CD34 + × 10 6 /kg infused was 10.35 (range 2– 50.2) (P = 0.0014). Patients in both groups were trans- planted with a variety of high-dose conditioning regimens not including total body irradiation. No patient required steroid therapy for regimen-related toxicities. Thirty-two patients after CD34-selected PBSC transplantation received G-CSF 5 g/kg post transplant while in the group of patients submitted to unfractionated PBSC, G-CSF was not used as for institutional policy. All patients who were sero- logically positive for herpes simplex virus (100% in both groups) received acyclovir 500 mg/m 2 every 8 h from day -1 to day 100 after transplantation. Patients were switched Table 1 Characteristics of patients Sex 17F/23M Median age (years) 40.5 (range 17–59) Disease (No. of patients) NHL 15 HD 12 MM 12 RMS 1 Conditioning regimen BuMel 18 BEAM 11 BuCy2 6 TTMel 2 BuTTMel 1 Median No. of CD34 + reinfused 4.2 × 10 6 /kg (range 0.86–16) Median day of PMN 500/l + 13 (range 8–17) Median day of platelets 50 000/l + 21 (range 10–120) Median day of reticulocytes 1% + 15 (range 11–105) Median No. of RBCu 1 (range 0–16) Median No. of SDU 2 (range 0–8) Disease status (No. of patients) CR 19 PD 9 PR 6 Rel 5 Lost FU 1 Survival status (No.) Alive 32/Dead 8 Median overall survival post-transplant 23 (range 1–67) (months) TT = thiotepa; RBCu = red blood cell unit; SDU = single donor unit; FU = follow up. to oral high-dose acyclovir as soon as possible after auto- transplantation. Antiviral therapy with gancyclovir or fos- carnet was administered for documented CMV infection. Treatment with cidofovir or ribavirin was instituted when a documented adenovirus infection was diagnosed. No significant differences were observed for neutrophil engraftment in both groups (13 vs 13 days) while platelet (21 vs 12 days, P = 0.0001) and reticulocyte (15 vs 12, respectively, P = 0.0001) engraftment were significantly delayed after CD34-selected PBSC transplantation produc- ing a significantly higher requirement for platelet and RBC transfusions. The absolute number of lymphocytes and the count of CD4 + lymphocytes was significantly lower at 30 and 60 days in CD34-selected patients as reported elsewhere. 4 Five patients (12.5%) developed CMV infection between days + 4 to + 28 after infusion of CD34-selected PBSC, one patient (2.5%) developed CMV enteritis and five additional patients (12.5%) developed adenovirus infection (hemorrhagic cystitis or enteritis). No patient developed CMV pneumonia and no CMV or virus-related deaths were encountered. In the group of patients submitted to unfractionated PBSC, CMV dis- ease and infection were absent and adenovirus or other viruses were not isolated within the first 100 days after reinfusion. The incidence of CMV infection and disease was sig- nificantly higher in our CD34-selected patients as well as the incidence of adenovirus-associated disease (P = 0.0001 and P = 0.005, respectively). We do agree with the data showed by Holmberg et al and the data altogether are con-