Introduction

Hematopoietic stem cell transplantation (HSCT) using peripheral blood (PB) or bone marrow (BM) has become the standard treatment for a variety of both congenital and acquired hematologic disorders. Despite these recent advances, HSCT continues to be associated with significant risk and mortality resulting from the profound immunosupression induced by preparative regimens that are often myeloablative. Further, graft-versus-host disease (GVHD) and superimposed infections frequently compromise the care of these patients. As a result of these serious complications, they are often transferred to the intensive care unit (ICU).¹

Recently, umbilical cord blood (UCB) has emerged as an alternate source of donor hematopoietic stem cells.^{2, 3} The increasing use of UCB is driven by the fact that more than two thirds of patients awaiting HSCT lack a suitable histocompatible related donor and <50% of unrelated donor searches identify a suitably matched and available BM or PB stem cell graft. Several studies in both adult and pediatric patients have analyzed the outcomes of BMT and PBSCT patients requiring ICU expertise or mechanical ventilation.^{4, 5, 6, 7, 8} However, adult UCBT recipients may have different critical care needs and outcomes than BM and PB transplant patients due to different preparative regimens and infusion of lower numbers of hematopoietic stem cells than BM or PB recipients.⁹ To our knowledge, there have been no published studies of adult patients receiving UCBT and requiring ICU support. The purpose of the present study was to identify specific factors in UCBT patients (demographic, physiologic, and clinical) as potentially important predictors for ICU admission, as well as predictors of ICU outcome.

Methods

Data were collected on all patients who underwent UCBT at University Hospitals of Cleveland (UHC) between 1 January 1998 and 31 December 2003 under a study protocol approved by the Case Western Reserve University, Institutional Review Board. Patients were identified using an Ireland Cancer Center, UHC database. The methods for UCBT have been described previously.^{9, 10, 11} Patient eligibility included acute myeloid and acute lymphoblastic leukemia patients in complete remission, chronic myeloid leukemia patients in chronic or accelerated phase, non-Hodgkin's and Hodgkin's disease patients in second or subsequent second complete remission or responding relapse, aplastic anemia and high-risk myelodysplastic syndrome patients. Total body irradiation for myeloablative preparative regimens was 1200 cGy and for non-myeloablative regimens was 200 cGy.

For patients with more than one ICU admission only the first admission was considered to maintain independence of the observations. All physiologic and clinical data were collected at the time of admission to the ICU. Microbiologic cultures positive within 48 h of ICU admission were considered to reflect an infection at the time of ICU admission. The UHC inpatient Blood and Marrow Transplant Unit attending physician or the patient's primary transplant physician requested transfer of all patients included in this study to the MICU. All UCBT patients transferred to the MICU were managed by the MICU team (critical care attending physician, a critical care Fellow, four to five resident trainees, a nurse practitioner) with Bone Marrow Transplant team consultation.

Respiratory failure was defined as arterial P_CO2>50 mm Hg, or P_O2 <100 mm Hg with a fraction of inspired oxygen above 0.35, impending respiratory failure requiring intubation, the use of non-invasive ventilation, or the use of high-flow oxygen. Mechanical ventilation was considered as machine assisted tidal breathing through an endotracheal tube. Cardiovascular failure was defined by the use of vasopressor (dopamine >5 g/kg/min, noradrenaline, vasopressin, phenylephrine, adrenaline) infusions for more than 4 h. Acute renal failure was defined as an increase in serum creatinine concentration more than 50%, or 0.5 mg/dl above baseline value (recorded at least 6 months before transplant). Hemodialysis included continuous and intermittent modalities. Hepatic failure was defined as a total serum bilirubin >4 mg/dl. Acute GVHD was graded I–IV, based on the published grading system.^{12, 13} Acute Physiologic and Chronic Health Evaluation (APACHE) III score was derived on ICU admission.¹⁴ Neutropenia was defined as <500 neutrophils/l.

The significance of variables predicting transfer to, or death in, the ICU was determined by Fisher's exact test for dichotomous variables, and by Wilcoxon rank-sum test for continuous variables. Results were considered significant at an alpha error <0.05, two-tailed. Analyses were performed using SAS version 9.1 for Windows (SAS Institute Inc., Cary, NC, USA).

Results

Study population

Forty-four patients underwent UCBT from 1 January 1998 to 31 December 2003 and all were included in the analysis (Table 1). UCBT began at University Hospitals of Cleveland in 1998 and continues to the present time. The median patient age was 39 and varied from 16 to 69 years. Of UCBT patients, 48% were female. The number of transplants were uniformly distributed over time with the exception of 1999 when twice as many were performed in comparison to other years. The most common indication for UCBT was acute leukemia. Half of the patients had active disease at the time of transplant. Patients with chronic leukemia, lymphoma, or other diseases had responding yet evident disease, while all acute leukemia patients were in remission. Patients (80%) underwent a myeloablative preparative regimen and 41% of the patients received total body irradiation as part of the preparative regimen. The umbilical cord blood cell dose given varied over a sevenfold range with a median of 2.20 10⁷ nucleated cells infused/kg body weight.

Table 1 - Umbilical cord blood transplant study population.

Full table

Predictors of ICU transfer

Twenty-five patients (57%) were transferred to the ICU at some time after their transplant. ICU transfer occurred at a median of 42 days after transplant and ranged from 10 to 670 days. Diagnoses resulting in ICU admission were pneumonia (52%), lower gastrointestinal bleeding (12%), sepsis (8%), renal failure (8%) and other causes (20%) including ventilatory failure, cardiac arrest, aspiration pneumonitis and neurological deterioration, each of which accounted for one patient (Table 2). At the time of ICU admission 64% of patients had respiratory failure of which 48% required mechanical ventilation, 52% had cardiovascular failure, 80% had renal failure, 59% had liver dysfunction, 84% had thrombocytopenia and 52% had neutropenia. Identified infectious pathogens associated with ICU transfer included Streptococcus pneumonia isolated from blood cultures of one patient, Pseudomonas aeruginosa in the blood of one patient, vancomycin resistant Enterococcus faecium in the urine of one patient, CMV from the lung lavage of one patient, Aspergillus from the skin of one patient and Pseudollessheria boydii from the skin of one patient. The mean APACHE III score of all patients at the time of ICU admission was 92 (s.d.=30; range=38–145).

Table 2 - Diagnoses on ICU admission.

Full table

As shown in Table 3, the use of a myeloablative preparative regimen was significantly associated with ICU transfer (P=0.03) and total body irradiation trended towards the need for ICU admission (P=0.06). The presence of any GVHD, as well as CMV was not associated with ICU transfer in our small cohort. An umbilical cord blood cell dose over the median for our cohort (2.2 10⁷ nucleated cells/kg body weight) trended toward a protective effect on ICU transfer (P=0.07), as did male gender (P=0.08). A significant protective effect from ICU transfer was found with higher cell dose (P=0.05). Over time, our experience has shown a decreasing need for ICU transfer when compared to our initial year (P=0.01).

Table 3 - Variables associated with ICU transfer.

Full table

Predictors of ICU mortality

Once transferred to the ICU, the ICU mortality of UCBT patients was 72% (18 of 25 patients). Mortality was significantly associated with an ICU admission diagnosis of pneumonia (P=0.03). UCBT patients hospitalized, but not requiring ICU care had a 36% overall hospital mortality (data not shown). For UCBT patients who were transferred to the ICU, the median length of stay was 5 days with a range of 1–30 days. Table 4 outlines the results of continuous and categorical variables analyzed as predictors of ICU mortality at the time of admission. Significant factors were the use of vasopressor support (P=0.03), increasing APACHE III score (P=0.0004), and platelet count (P=0.03). A shorter time between transplant and ICU transfer was also suggestive of increased ICU mortality (P=0.09). Factors which were not significant predictors of ICU mortality were gender, disease category, disease status at the time of transplant, CMV status, chemotherapy regimen, the use of total body irradiation, cell dose, GVHD, respiratory failure, the need for mechanical ventilation, renal failure, liver dysfunction, age and infused cell dose.

Table 4 - Variables associated with ICU death.

Full table

In all 28% (seven patients) were discharged from the ICU after an average length of stay of 3.72.2 days (Table 5). All were ultimately discharged home with a median survival after ICU discharge of 113 days (mean=325575.4 days). In comparison, UCBT patients that did not require ICU care had a median survival of 855 days (mean=947764.5 days) post transplant.

Table 5 - Survival post-transplant by ICU status.

Full table

Discussion

To our knowledge, this is the first study to analyze outcomes of adult UCBT patients requiring ICU care. Several variables are notable in this analysis that may be useful in caring for this patient population as well as providing insight into variables that may be modified in the UCBT process in the drive for improved outcomes. First, the need for ICU care in this patient population is high. In our transplant population, 25 of 44 (57%) patients required ICU admission at some time. The majority required ICU admission within 90 days of transplant. The use of a myeloablative preparative regimen and total body irradiation, to some degree, were associated with the eventual need for ICU care. It is not clear from our analysis why these regimens are associated with ICU transfer. However, their use in preparation for UCBT may lead to more severe bone marrow suppression and visceral organ damage. Previous studies have shown that adult UCBT recipients have a longer interval to recovery of marrow function than BM or PB recipients,^{9, 10, 11} due, in part, to infusion of lower hematopoietic stem cell dose. During this recovery phase, these patients are at risk for bleeding complications, anemia resulting in impaired oxygen delivery, an increased susceptibility to opportunistic infections and sepsis. These complications alone, as well as any progression to multi-organ system failure, result in a need for ICU care. Importantly, in our analysis higher infused cell doses were associated with a significantly decreased need for ICU transfer. Several studies have shown that a higher nucleated cell content of the umbilical cord blood graft decreases the time to myeloid recovery and increases the probability of successful engraftment.^{15, 16, 17} Therefore, lower cell doses in our patients may have resulted in poor engraftment, subsequent acute organ dysfunction, and an increased complication rate. Multiunit UCBT from partially matched unrelated donors has been used to increase the number of nucleated cells infused and may be a method to improve myeloid recovery and patient outcome.¹⁸

No prior studies in adults with UCBT are available to compare the results of our ICU transfer and mortality. However, prior studies of hematopoietic stem cell transplantation are available for comparison, predominantly in patients who require mechanical ventilation.¹⁹ In this patient population mortality rates range from 60 to 98% (19). In our UCBT cohort, the mortality in patients requiring mechanical ventilation was 83 and 62% in the non-mechanically ventilated patients. Thus, the UCBT survival information is roughly comparable to mortality rates seen in hematopoietic stem cell transplantation. Prediction of ICU mortality included hemodynamic instability (vasopressor use), lower platelet count, and higher APACHE III score, with GVHD and duration of UCBT trending toward significance. Hemodynamic instability has been noted to predict ICU mortality in BMT and PBSCT pediatric and adult patients.^{4, 20, 21, 22, 23} As part of the hemodynamic instability, tachycardia was also associated with mortality. Tachycardia was a more sensitive predictor of mortality than mean or systolic blood pressure. However, the utility of tachycardia as a prognostic factor suffers from its lack of specificity. Its utility may lie in heightening the awareness of the care team for possible hemodynamic failure and the need for ICU support.

Thrombocytopenia also predicted ICU mortality. Thrombocytopenia may reflect lack of stem cell engraftment, bone marrow failure, bleeding complications or infection. Platelet engraftment status on ICU admission and bleeding complications are known significant predictors of ICU survival in HSCT.^{5, 6, 24} In the UCBT population, thrombocytopenia may reflect incomplete or slower stem cell engraftment and its association with mortality. A shorter time period between UCBT and ICU transfer trended toward statistical significance as a predictor of ICU mortality, and has also been recognized in prior analyses.^{22, 24} The frequent pancytopenia that occurs in the early post-UCBT period with impaired oxygen delivery, infectious complications, sepsis and multi-organ system failure, puts patients at significant risk of mortality.

The APACHE III score was a very sensitive and specific predictor of ICU mortality in our analysis. Several other reports in BMT and PBSCT patients also have identified higher APACHE II and III scores as a significant predictor of mortality.^{6, 21, 23, 24, 25, 26} In the current analysis, no patient survived with an APACHE III score on ICU admission >89. This suggests that patients may be transferred to the ICU late in their clinical course with severe physiologic derangement and raises the question of whether these patients would benefit from early ICU transfer, even with signs of minimal organ dysfunction. Although not validated in non-ICU patients, APACHE III score or an alternate simplified physiologic scoring system might be used to monitor the severity of illness in UCBT patients to aid in decision-making regarding ICU transfer before the onset of multi-organ system failure.

Parameters not associated with ICU mortality included; age, gender, disease category, disease status at the time of transplant, CMV status, chemotherapy regimen, total body irradiation, umbilical cord blood stem cell dose, blood pressure, and renal, hepatic or respiratory failure. We did not identify mechanical ventilation on ICU admission as a significant predictor of mortality in our population. Prolonged mechanical ventilation has been shown to be a significant mortality predictor in the majority of both adult and pediatric BMT and PBSCT. The cohort size will need to be expanded to verify this finding.

The limitations of our study include its single-center nature and a small population limiting its power and the ability to draw general conclusions. However, these data provide important directions for future analysis including a direct comparison of non-myeloablative to myeloablative preparative regimens, analysis of regimens with or without total body irradiation and the impact of cell dose on outcome. Furthermore, a prospective analysis of APACHE III scores or a simplified physiology of organ failure score, as a tool to determine the need for critical care in this patient population is warranted. Importantly, the extent to which early identification of patients requiring initiation of intensive care results in improved outcome should be studied. Finally, mortality remains significant in this ICU population. These data should be used to guide expectations of patients, families and health care delivery team members until new approaches to patient support and treatment are defined.

References

1. Naeem N, Reed MD, Creger RJ, Youngner SJ, Lazarus HM. Transfer of the hematopoietic stem cell transplant patient to the intensive care unit: does it really matter? Bone Marrow Transplant 2006; 37: 119–133. | Article | PubMed | ChemPort |
2. Gluckman E, Broxmeyer HA, Auerbach AD, Friedman HS, Douglas GW, Devergie A et al. Hematopoietic reconstitution in a patient with Fanconi's anemia by means of umbilical cord blood from an HLA-identical sibling. N Engl J Med 1989; 321: 1174–1178. | PubMed | ISI | ChemPort |
3. Wagner JE, Broxmeyer HE, Byrd RL, Zehnbauer B, Schmeckpeper B, Shah N et al. Transplantation of umbilical cord blood after myeloablative therapy: analysis of engraftment. Blood 1992; 79: 1874–1881. | PubMed | ISI | ChemPort |
4. Rubenfeld GD, Crawford SW. Withdrawing life support from mechanically ventilated recipients of bone marrow transplants: a case for evidence based guidelines. Ann Intern Med 1996; 125: 625–633. | PubMed | ISI | ChemPort |
5. Price KJ, Thall P, Kish SK, Shannon VR, Anderson BS. Prognostic indicators for blood and marrow transplant patients admitted to an intensive care unit. Am J Resp Crit Care Med 1998; 158: 876–884. | PubMed | ISI | ChemPort |
6. Afessa B, Tefferi A, Dunn WF, Litzow MR, Peters SG. Intensive care unit support and acute physiology and chronic health evaluation III performance in hematopoietic stem cell transplant recipients. Crit Care Med 2003; 31: 1715–1721. | Article | PubMed | ISI |
7. Soubani AO, Kseibi E, Bander JJ, Klein JL, Khanchandani G, Ahmed HP et al. Outcome and prognostic factors of hematopoietic stem cell transplantation recipients admitted to a medical ICU. Chest 2004; 126: 1604–1611. | Article | PubMed | ISI |
8. Hagen SA, Craig DM, Martin PL, Plumer DD, Gentile MA, Schulman SR et al. Mechanically ventilated pediatric stem cell transplant recipients: effect of cord blood transplant and organ dysfunction on outcome. Pediat Crit Care Med 2003; 4: 206–213. | Article |
9. Laughlin MJ, Eapen M, Rubinstein P, Wagner JE, Zhang M-J, Champlin RE et al. Outcomes after transplantation of cord blood or bone marrow from unrelated donors in adults with leukemia. N Engl J Med 2004; 351: 2265–2275. | Article | PubMed | ISI | ChemPort |
10. Laughlin MJ, Barker J, Bambach B, Koc ON, Rizzieri DR, Wagner J et al. Hematologic engraftment and survival after unrelated placental cord transplant in adult recipients. N Engl J Med 2001; 344: 1815–1822. | Article | PubMed | ISI | ChemPort |
11. Hamza NS, Lisgaris M, Yadavalli G, Nadeau L, Fox R, Fu P et al. Kinetics of myeloid and lymphocyte recovery and infectious complications after unrelated umbilical cord blood versus HLA-matched unrelated donor allogeneic transplantation in adults. Br J Haematol 2004; 124: 488–498. | Article | PubMed |
12. Przepiorka D, Weisdorf D, Martin P, Klingemann HG, Beatty P, Hows J et al. 1994 Consensus Conference on Acute GVHD Grading. Bone Marrow Transplant 1995; 15: 825–828. | PubMed | ISI | ChemPort |
13. Flowers ME, Kanasu E, Sullivan KM. Pathophysiology and treatment of graft-versus-host disease. Hematol Oncol Clin North Am 1999; 13: 1091–1112. | Article | PubMed | ChemPort |
14. Knaus WA, Wagner DP, Draper EA, Zimmerman JE, Bergner M, Bastos PG et al. The APACHE III prognostic system: risk prediction of hospital mortality for critically ill hospitalized adults. Chest 1991; 100: 1619–1636. | PubMed | ChemPort |
15. Gluckman E, Rocha V, Boyer-Chammand A, Locatelli F, Arcese W, Pasquini R et al. Outcome of cord-blood transplantation from related and unrelated donors: Eurocord Transplant Group and the European Blood and Marrow Transplantation Group. N Eng J Med 1997; 337: 373–381. | Article | ChemPort |
16. Rubinstein P, Carrier C, Scaradavou A, Kurtzberg J, Adamson J, Migliaccio AR et al. Outcomes among 562 recipients of placental-blood transplants from unrelated donors. N Engl J Med 1998; 339: 1565–1577. | Article | PubMed | ISI | ChemPort |
17. Wagner JE, Barker JN, DeFor TN, Baker KS, Blazar BR, Eide C et al. Transplantation of unrelated donor umbilical cord blood in 102 patients with malignant and nonmalignant diseases: influence of CD34 cell dose and HLA disparity on treatment-related mortality and survival. Blood 2002; 100: 1611–1618. | PubMed | ISI | ChemPort |
18. Barker JN, Weisdorf DJ, Wagner JE. Creation of a double chimera after the transplantation of umbilical-cord blood from two partially matched unrelated donors. N Engl J Med 2001; 344: 1870–1871. | Article | PubMed | ISI | ChemPort |
19. Bach PB, Schrag D, Nierman DM, Horak D, White Jr P, Young JW et al. Identification of poor prognostic features among patients requiring mechanical ventilation after hematopoietic stem cell transplantation. Blood 2001; 98: 3234–3240. | Article | PubMed | ISI | ChemPort |
20. Hollmig KA, Soehngen D, Leschke M, Kobbe G, Schneider P, Klein RM et al. Long-term survival of recipients of allogeneic bone marrow transplantation after mechanical ventilation. Eur J Med Res 1997; 2: 62–66. | PubMed | ChemPort |
21. Jackson SR, Tweeddale MG, Barnett MJ, Spinelli JJ, Sutherland HJ, Reece DE et al. Admission of bone marrow transplant recipients to the intensive care unit: outcome, survival and prognostic factors. Bone Marrow Transplant 1998; 21: 697–704. | Article | PubMed | ISI | ChemPort |
22. Shorr AF, Moores LK, Edenfield WJ, Christie RJ, Fitzpatrick TM. Mechanical ventilation in hematopoietic stem cell transplantation. Chest 1999; 116: 1012–1018. | Article | PubMed | ISI | ChemPort |
23. Scott PH, Morgan TJ, Durrant S, Boots RJ. Survival following mechanical ventilation of recipients of bone marrow transplants and peripheral blood stem cell transplants. Anaesth Intensive Care 2002; 30: 289–294. | PubMed | ISI | ChemPort |
24. Chuk W, Wong KO, Wong CS, Chan JK. Prognostic factors in children requiring admission to an ICU after hematopoietic stem cell transplant. Hematol Oncol 2004; 22: 1–9. | Article | PubMed | ISI |
25. Afessa B, Tefferi A, Hoagland HC, Letendre L, Peters SG. Outcome of recipients of bone marrow transplants who require intensive-care unit support. Mayo Clinic Proc 1992; 67: 117–122. | ISI | ChemPort |
26. Paz HL, Crilley P, Weinar M, Brodsky I. Outcome of patients requiring medical ICU admission following bone marrow transplantation. Chest 1993; 104: 527–531. | PubMed | ISI | ChemPort |

Acknowledgements

We thank Shawn Knapik and Er-Wey Teng for initial discussions and project initiation and Robert Fox for providing patient transplant data. Logistic support and patient data were provided by the Ireland Cancer Center and the Medical Records Department at the University Hospitals of Cleveland. Dr Elizabeth Kern was supported as a VA National Quality Scholar.