Robert I. Parker

Thrombosis in the pediatric population.

Whereas thrombotic events in critically ill children do not occur as commonly as in adults, they are being recognized with increasing frequency in the pediatric intensive care unit. The reasons for this are not clear but likely include an increased awareness of the problem and the ability to make a diagnosis using relatively noninvasive tests. In this section, I attempt to define the extent of the problem, summarize and discuss the relevant literature (pointing out where published experience in the pediatric population differs from that in adult patients), and suggest some guidelines regarding thrombophilia treatment and the management of thrombotic events.

Transfusion in critically ill children: indications, risks, and challenges.

Objective:To provide a concise review of transfusion-related issues and practices in the pediatric patient population, with a focus on those issues of particular importance to the care of critically ill children. Data Source:Electronic search of the PubMed database using the search terms “pediatric transfusion,” “transfusion practices,” “transfusion risks,” “packed red blood cell transfusion,” “white blood cell transfusion,” “platelet transfusion,” “plasma transfusion,” and “massive transfusion” either singly or in combination. Study Selection and Data Extraction:All identified articles published since 2000 were manually reviewed for study design, content, and support for indicated conclusions, and the bibliographies were scrutinized for pertinent references not identified in the PubMed search. Selected studies from this group were then manually reviewed for possible inclusion in this review. Data Synthesis:Well-designed studies have demonstrated the benefit of “restrictive” transfusion practices across the entire age spectrum of pediatric patients across a wide spectrum of pediatric illness. However, clinician implementation of the more restrictive transfusion practices supported by these studies is variable. Additionally, the utilization of both platelet and plasma transfusions in either a “prophylactic” or “therapeutic” setting appears to be greater than that supported by published data. Conclusions:The preponderance of prospective, randomized trials and retrospective analyses support the use of a restrictive packed RBC transfusion policy in most clinical conditions in children. Neonatal transfusions guidelines rely largely on “expert opinion” rather than experimental data. Current transfusion practices for both platelets and coagulant products (e.g., fresh-frozen plasma and recombinant-activated factor VII) are poorly aligned with recommended transfusion guidelines. As with adults, current transfusion practices in children often do not reflect implementation of our current knowledge on the need for transfusion. Greater efforts to implement current evidence-based transfusion practices are needed.

Thrombosis in the intensive care unit: protein C and a whole lot more!

Hemostasis is a dynamic process wherein the balance between clot formation and inhibition of coagulation are tightly controlled to stop unwanted bleeding and prevent unwanted thrombosis. Although hemorrhage has long been recognized as a major factor contributing to the morbidity and mortality of critically ill patients, pathologic thrombosis, other than that which occurs in acute coronary syndromes, pulmonary emboli, and postoperative deep venous thrombosis, has only recently received the attention it deserves in this patient population. This may be, in part, attributable to the fact that only recently has medical evaluation been able to identify gene and protein abnormalities that provide insight into the pathophysiology of thrombi formation in those patients who display a thrombophilic phenotype. However, we cannot discount the possibility that this lack of attention is also attributable in part to the fact that thrombosis, especially thrombosis at the microvascular level, is often silent. In contrast to bleeding, the clinician cannot directly see or measure thrombi formation until it has progressed to the point that organ or limb function has been compromised. By that time, the patient’s prognosis is often compromised as well. Over the years, a subset of intensivists and hematologists have put forth the hypothesis that multiorgan failure is a consequence of microvascular thrombosis; consequently, there has been a steady stream of bench and clinical progress in this area. The Margaux conference of 2001 addressed this issue in the context of sepsis/septic shock (1–4). Whereas many trials of novel proteins have failed to show clinical benefit (5, 6), the “high water” mark in this regard came with the publication of the results of the PROWESS and ENHANCE US trials by Barnard et al (7–9). The fact that activated protein C was chosen as a target therapy for microvascular thrombosis in sepsis is not surprising given the large body of laboratory data indicating the central role activated protein C plays in the interaction between inflammation and coagulation (10–12). One of this issue’s contributions is an insightful review of the role the protein C system plays in the regulation of hemostasis. Although recombinant human-activated protein C has been shown to be beneficial in selected patients, the initial enthusiasm after the publication of the PROWESS data has not been universally sustained because of side effects (bleeding), cost, efficacy, and difficulty in identifying appropriate patients who would best benefit from this therapy (13–16). What may have been lost in all of the discussion and controversy surrounding the use of recombinant activated protein C in sepsis is the fact that over the past several years, our knowledge of how hemostasis is regulated has improved and other clinically relevant risk factors for thrombosis have been described. These factors include gene mutations/polymorphisms, abnormalities in the blood levels of specific proteins (too high or too low), and acquired abnormalities (17–23). Although the R506Q mutation in the factor V gene (factor V Leiden) identified by the Dutch group (in Leiden) may be the most widely known of these risk factors, it is only one of several such risk factors. For many of these factors, the strength of the venous thromboembolism risk is marginal or only putative. A listing of some of these risk factors along with a suggested priority for testing in intensive care unit (ICU) patients is shown in the Table 1. There is no consensus regarding the appropriate hematologic workup of a patient who presents with a pathologic thrombus or in whom a pathologic thrombus develops, and “experts” can disagree as to just what an initial evaluation should include when looking for underlying risk factors. In general, it is always prudent to tailor one’s evaluation to the specific clinical picture presented by the patient. However, it is anticipated that a better understanding of the basis for pathologic thrombus formation will translate to better decisions in this regard. Although much has been written concerning the value, or lack thereof, of testing patients who experience an “unprovoked venous thromboembolism, and their relatives” (22–26), even in the presence of consensus guidelines for thrombophilia testing, studies have shown that appropriate patients are frequently not tested whereas others are inappropriately tested for the presence of thrombotic risk factors (27, 28). When confronted with a critically ill patient who manifests a clinically significant thrombotic event or process, testing at a minimum should include those elements that we can “fix,” i.e., the replacement of deficient anticoagulant proteins that will damp-down the coagulation process (e.g., protein C, protein S, antithrombin III). The finding of a genetic risk (e.g., factor V Leiden) will not change one’s initial management but may affect therapy after the resolution of the acute illness (29). Application of the findings and recommendations derived from population studies to patients in the ICU should be attempted with caution. One could argue that any patient in the ICU has an increased risk for thrombosis because that patient is in the ICU. This risk accrues from both the condition that From Pediatric Hematology/Oncology, Stony Brook University Cancer Center, SUNY at Stony Brook, Stony Brook, NY. The author has not disclosed any potential conflicts of interest. For information regarding this article, E-mail: robert.parker@stonybrook.edu Copyright

Platelet Transfusions in the PICU: Tiny Cells, Big Issue.

Pediatric Critical Care Medicine www.pccmjournal.org 897 the interim, efforts to continually improve the performance of POCT glucometers as well as strategies to reduce user error and variance in sampling are of paramount importance. Using the example of tight glucose control in children with severe burns, this study demonstrates that point-of-care BG monitoring can reach the sweet spot without getting everyone hot and bothered!

RBC transfusion practices: once again, we have met the enemy and they are us!

www.ccmjournal.org 2449 Over the years, the critical hemoglobin that triggers a decision to transfuse RBCs has changed as we have gained a better understanding of the benefits and risks of RBC transfusion. With the publication of the study by Hébert et al (1) demonstrating that a restricted transfusion strategy does not result in a worse outcome for most ICU patients, and subsequent publication of studies demonstrating similar results for both pediatric (2) and neonatal patients (3), there appears to be a consensus supporting a restrictive transfusion approach for most critically ill patients except for those in selected patient groups with trauma, neuroinjury, or acute coronary syndromes (4–7). The issue with which we are now presented involves the actual implementation of a restricted transfusion policy in otherwise “stable” critically ill patients. Imbedded in this question is the issue of how new treatment practices become implemented by clinicians and over what time frame are these changes made. In this issue of Critical Care Medicine, Murphy et al (8) present a retrospective study in which they attempted to measure changes in transfusion practices over a time interval during which the major studies validating the safety of restrictive transfusion practices were published. The data were analyzed for the type, size, and teaching status of the ICUs using data submitted to a state-wide registry. The authors found an overall decrease in transfusions administered to ICU patients across the entire time period, although the decrease was limited to large ICUs (defined as having > 200 discharges/yr). Small ICUs (< 200 discharges/yr) did not demonstrate a similar decrease in transfusion prevalence. This difference could be explained by unit or physician experience, that is, physicians who see more patients more readily become comfortable with a restrictive transfusion program. Although one might also expect ICUs in teaching hospitals to more likely be early adapters of a restrictive transfusion policy, the teaching status of a unit did not appear to affect transfusion practices even though more of the large ICUs were in teaching hospitals compared with the small units (42% vs 30%). This latter point is somewhat surprising as one might expect physicians in teaching hospitals to be more likely to be comfortable implementing new care paradigms. However, the study by Murphy et al (8) does not appear to reflect this expected physician behavior. If we in medicine who teach residents and fellows do not implement new (and presumably better) care paradigms, then who will? So-called academic physicians have repeatedly demonstrated a willingness to explore new ways to improve care through their research activities but have simultaneously been slow to adopt changes in care that represent a change from their own “standard care”—by this I mean changes that take us out of our comfort zone. Repeatedly, we have been slow to adopt guideline-directed care for medical conditions across virtually all medical specialties. In the ICU, these conditions include thromboembolism prevention, ventilator management, sepsis care, and management of patients with traumatic brain injury. This failure to adopt management guidelines occurs even when adoption has been shown to be beneficial, and nonadherence to guidelines results in increased mortality as has been shown with sepsis care (9– 11). Fortunately, the sepsis care guidelines have gained wide, although not universal, acceptance with the result being improvement in survival. Unfortunately, other treatment guidelines have not been so widely implemented. Why is this so? Studies investigating physician noncompliance have identified several factors that appear to affect adoption of guideline care. These include physician characteristics (personality traits, age, time since completion of training, gender, ethnic/ racial origin), patient comorbidities, complexity of guideline algorithm, institutional support, and education directed toward the targeted therapy (12–15). So, if the medical community is willing to implement some guidelines but is resistant to implement others, why has a restricted transfusion regimen not gained wider acceptance? Possibly (an hypothesis) RBC transfusion guidelines have not been aggressively accepted and implemented because those nonadopters do not believe the risks of RBC transfusion are sufficient to justify a reduction in RBC exposure, or possibly the benefit of decreased RBC exposure on outcome is not “real” enough to them (e.g., the patient is perceived as dying as a direct consequence of the transfusion in contrast to the case with sepsis and sepsis guidelines). If this is the case, then the task is to implement education programs to change attitudes. This has been shown to be an important component of most successful programs and can result in a decrease in RBC transfusion when targeted to that clinical question (16– 18). The good news here is that although we are the enemy, we are also the solution to implementing more rational blood transfusion practices in the ICU. *See also p. 2344.

The big chill: cooling sickle cells with caution.

1. Annane D, Vignon P, Renault A, et al: Norepinephrine plus dobutamine versus epinephrine alone for management of septic shock: A randomised trial. Lancet 2007; 370: 676–684 2. De Backer D, Biston P, Devriendt J, et al: Comparison of dopamine and norepinephrine in the treatment of shock. N Engl J Med 2010; 362:779–789 3. Sorsa T, Pollesello P, Solaro RJ: The contractile apparatus as a target for drugs against heart failure: Interaction of levosimendan, a calcium sensitiser, with cardiac troponin c. Mol Cell Biochem 2004; 266:87–107 4. van Hees HW, Dekhuijzen PN, Heunks LM: Levosimendan enhances force generation of diaphragm muscle from patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2009; 179: 41– 47 5. Doorduin J, Sinderby CA, Beck J, et al: The calcium sensitizer levosimendan improves human diaphragm function. Am J Respir Crit Care Med 2011 Sep 29. [Epub ahead of print] 6. Landoni G, Mizzi A, Biondi-Zoccai G, et al: Levosimendan reduces mortality in critically ill patients. A meta-analysis of randomized controlled studies. Minerva Anestesiol 2010; 76:276–286 7. Landoni G, Biondi-Zoccai G, Greco M, et al: Effects of levosimendan on mortality and hospitalization. A meta-analysis of randomized controlled studies. Crit Care Med 2012; 40:634–646 8. Mebazaa A, Nieminen MS, Packer M, et al: Levosimendan vs dobutamine for patients with acute decompensated heart failure: The SURVIVE Randomized Trial. JAMA 2007; 297:1883–1891 9. Packer M: Revive II: Multicenter placebocontrolled trial of levosimendan on clinical status in acutely decompensated heart failure [Abstract]. Circulation 2005; 112:3363 10. Task Force for Diagnosis and Treatment of Acute and Chronic Heart Failure 2008 of European Society of Cardiology, Dickstein K, Cohen-Solal A, et al: ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2008: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2008 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur Heart J 2008; 29: 2388 –2442

Pediatric intensive care unit outcome following pediatric hematopoietic stem cell transplantation: quo vadis?

Not long ago, a respected colleague at my hospital asked if our intensive care unit (ICU) had a weaning protocol. I was shocked because we had one for 5 yrs, developed by an enthusiastic team, following published guidelines (1). Although we routinely measure performance, we do not monitor adherence to our weaning protocol. If asked, I would have assumed naively that our practice was excellent and the protocol was followed more closely than it really was. Lesson learned: if you do not track performance, you do not really know what is happening in your unit. The frustrating gap between recommended and actual practice is particularly poignant for sepsis management. When used appropriately, the life-saving potential of individual therapies is staggering: a 16% absolute risk reduction for early goal directed therapy (2), 9% for low tidal volume ventilation (3), 17% for bacteremic patients managed with tight glucose control (4), 10% for low-dose corticosteroids (5), and at least 6% for recombinant activated protein C (6). The impressive math once prompted Dr. Gordon Bernard to suggest at a national meeting that we would soon have more patients leaving the ICU than going in. Although sepsis case fatality rates are decreasing (7), morbidity and mortality remain too high and recommended lifesaving therapies are used too inconsistently (8, 9). In this issue of Critical Care Medicine, Brunkhorst et al. (10) add to a growing list of studies describing the failure to employ evidence-based practice. On a single day in 2003, the investigators studied a random sample of 214 German ICUs, representing a wide range of hospitals, from small community institutions to large ones affiliated with universities. ICU directors were asked how often several evidence-based interventions were used, including low tidal volume ventilation for acute respiratory distress syndrome/acute lung injury, tight glycemic control, activated protein C, and low-dose hydrocortisone. Charts were audited to compare survey responses to actual practice. The authors studied 366 randomly selected patients with severe sepsis or septic shock. A total of 79.9% of ICU directors reported adherence to low tidal volume strategies. However, among patients with acute respiratory distress syndrome/acute lung injury, 80.3% audited were managed with volumes above 8 mL/kg predicted body weight. Similarly, 65.9% of ICU directors reported adherence to strict glucose control whereas 66.2% of patients audited were frankly hyperglycemic. Similar gaps between perceived and actual practice were shown for all strategies studied. Findings were the same for hospitals of all sizes and affiliations, although perceived adherence was higher in larger, academic institutions. This important study has many strengths. It was executed by well-respected investigators from the German Sepsis Competence Network (SepNet) and the quality of data collection was exemplary. The investigators studied a rich array of ICUs, and it is reasonable to consider the findings generalizable beyond Germany. The article does have limitations. Because the investigators focused on maximum tidal volumes and glucose measurements, adherence may seem worse than it really was: a single recording outside the recommended range constituted nonadherence, even if other measurements were acceptable. Use of average or median measurements might have been more convincing. In addition, the study was not designed to identify reasonable decisions to override guidelines, which must have occurred in some patients. Certainly 100% adherence to guidelines would be neither achievable nor desirable. Despite these concerns, it seems appropriate to conclude that practice shortcomings are widespread. Why do our colleagues fail to routinely employ evidence-based practice? Reasons include knowledge deficits, institutional barriers, and thoughtful disagreement (11–15). Strategies to promote adherence seem particularly suited to knowledge deficits and institutional barriers (15). For example, obstacles to using low tidal volumes include the failure to diagnose acute respiratory distress syndrome/acute lung injury and unwarranted concerns about sedation, both of which should be amenable to education (16, 17). Concerns over loss of autonomy (14) could be addressed by including all stakeholders when developing protocols. Potentially useful practices include standardized order sets, bundles, and easily followed algorithms (18–20). Reasoned disputes over guideline content raise a bigger challenge. Legitimate arguments can be made over the science supporting some recommendations (21, 22). The quality of guidelines and consensus statements has been criticized (12) and their content may fail to satisfy some experts (11). New findings can quickly undermine the rationale for certain recommendations such as aggressive glucose management (23). Little debate should remain over the value of low tidal volumes for acute respiratory distress syndrome/acute lung injury. Even allowing for subtle disagreements about study interpretation, consistent evidence supporting low tidal volumes (or against high ones) has essentially created a standard of care (3, 8, 24). Nevertheless, it is not difficult to imagine situations in which rigid adherence to a protocol would be scientifically unsound if not dangerous. For example, to avoid overdistention, the ARDSNet ventilation protocol recommends reducing volumes to as low as 4 mL/kg predicted body weight when plateau pressures exceed 30 cm H2O (3). However, external compression of the lungs, for instance due to abdominal compartment syndrome, could easily push plateau pres*See also p. 2719.