Risk factors for HL in pediatric ALL patients include age <1 year, T-cell phenotype, MLL (11q23) rearrangement, Philadelphia chromosome t(9;22) and leukemic cell ploidy ≤50 (Pui et al. 1990; Eguiguren et al. 1992; Vaitkevičienė et al. 2009). Historically, long-term outcomes in ALL patients with HL have been lowest, likely due to these underlying disease characteristics (Möricke et al. 2008; Vaitkevičienė et al. 2009; Gaynon et al. 2010). In ALL cooperative group studies, 10.9–13.1 % of patients had initial WBC ≥100 × 10⁹/L and 4.8–5.8 % had initial WBC ≥200 × 10⁹/L (Möricke et al. 2008; Gaynon et al. 2010; Lund et al. 2011). For the 132 patients with initial WBC ≥200 × 10⁹/L in the Scandinavian NOPHO ALL-92 and ALL-2000 protocols, 7 (5.3 %) suffered early death, 6 secondary to neurologic complications, and none secondary to TLS with 5 dying prior to therapy initiation (Lund et al. 2011).

In a follow-up study of 221 NOPHO patients with WBC ≥200 × 10⁹/L, Vaitkevičienė et al. (2013) importantly note that all complications secondary to HL, except the need for dialysis, occurred at significantly higher median WBC counts: neurologic (WBC 530 × 10⁹/L vs. 327 × 10⁹/L), respiratory (WBC 620 × 10⁹/L vs. 336 × 10⁹/L), bleeding (WBC 420 × 10⁹/L vs. 327 × 10⁹/L), and dialysis (WBC 310 × 10⁹/L vs. 357 × 10⁹/L). Neurologic and bleeding complications occurred at a significantly higher age, while respiratory distress occurred at a significantly younger age. In multivariate analysis, only WBC count and neurologic symptoms at admission were significantly associated with risk of early death; timing of administration of antileukemic therapy, admission hemoglobin level and administration of packed red blood cells (PRBCs) were not significant. Six patients (2.7 %) died within 2 weeks of presentation, all secondary to neurologic complications, with 4 of the 6 presenting with severe symptoms. Patients receiving mechanical cytoreduction (leukapheresis or exchange transfusion) prior to chemotherapy initiation had significant delay in antileukemic therapy initiation with no noted survival benefit. TLS occurred in 12 % of patients with statistically higher initial uric acid levels (11.0 vs. 7.7 mg/dL). Patients with T-cell and infant ALL were at increased risk of TLS, rasburicase significantly reduced the risk of TLS and no patient died from TLS. Initial uric acid level was the only significant factor for the development of TLS on multivariate analysis (WBC count, lactate dehydrogenase, corticosteroid dose and mechanical cytoreduction were not significant). In their supplemental HL guidelines, Vaitkevičienė et al. (2013) recommend prompt initiation of antileukemic therapy after diagnostic evaluations and rasburicase administration with no recommendation for leukapheresis or exchange transfusion.

Maurer et al. (1988) similarly reported an 8.4 % incidence of WBC >200 × 10⁹/L in a pediatric ALL cohort. Early death occurred in 7 of 124 (5.6 %) patients, 4 from sepsis, 1 from pneumonia, 1 from GI hemorrhage and 1 from ICH. Four (3.2 %) patients suffered ICH with 3 occurring at presentation and all with WBC >600 × 10⁹/L. Although electrolyte abnormalities were decreased in those undergoing leukapheresis, risk of renal dysfunction requiring dialysis was low, occurring in only 3 (2.4 %) patients, all of whom had received low-dose prednisone pretreatment. WBC >600 × 10⁹/L and massive splenomegaly were the only significant adverse prognostic factors on multivariate analysis.

In a comprehensive study of HL over a 40-year period at St. Jude Children’s Research Hospital, Lowe et al. (2005) reported 178 of 2,288 (7.8 %) children with ALL presenting with WBC >200 × 10⁹/L. Early death occurred in 7 (3.9 %), 3 from sepsis and 2 each from ICH and respiratory failure. Cytoreduction was performed in those with a median initial leukocyte count of 416 × 10⁹/L vs. 295 × 10⁹/L in the nonreduced cohort. The time to initiate chemotherapy was significantly longer in the cytoreduced cohort. Neurologic and respiratory complications were statistically more likely with WBC >400 × 10⁹/L. Older age was also significant for neurologic complications. Four patients suffered from ICH (2.2 %), all with initial WBC >400 × 10⁹/L and 3 at initial presentation to the tertiary care institution. Two of the 4 received PRBC transfusion prior to ICH, while the other 2 had initial hemoglobin of 10.1 and 10.2 g/dL. The majority of neurologic complications (13 of 16; 81 %) occurred at presentation; the additional 3 occurred 24 h after presentation, 1 in the cytoreduced group and 2 in the nonreduced cohort. Pulmonary complications occurred more commonly in the cytoreduced cohort and were equally common before and after cytoreduction. Metabolic complications were common, with hyperkalemia in 10 % and hyperphosphatemia in 20 %, but only 1 patient died secondary to metabolic dysfunction (hyperkalemia) and overt renal dysfunction was rare. Although it was unclear if cytoreduction was beneficial in the prevention of TLS or leukostasis symptoms and did not impact early mortality, Lowe et al. (2005) recommend cytoreduction for WBC >400 × 10⁹/L. Again, early initiation of chemotherapy in pediatric ALL seems most important although it is unlikely to impact those presenting with severe neurologic dysfunction and may not impact early death.

Like ALL, HL has been shown to be a negative prognostic factor in long-term outcome in pediatric AML patients (Inaba et al. 2008; Lange et al. 2008). In analysis of AML-BFM 93 and AML-BFM 98, Creutzig et al. (2004) reported a 19.1 % incidence of HL with significantly increased risk of early death with WBC ≥100 × 10⁹/L (9.9 % vs. 2.1 % with WBC <100 × 10⁹/L) and even more so with WBC ≥200 × 10⁹/L (16.9 % vs. 2.4 % with WBC <200 × 10⁹/L). Multivariate analysis found HL, FAB M5 and low-performance status as significant independent prognostic factors. A recent analysis of COG AML studies AAML03P1 and AAML0531 reported an 18.8 % incidence of HL with independent risk factors including age ≤1 year, FAB M1, M4, and M5, inv(16), and FLT3-ITD+ (Sung et al. 2012). These findings have been corroborated by other studies (Creutzig et al. 1987; Meshinchi et al. 2001; Inaba et al. 2008). TLS was rare although hyperphosphatemia and hyperuricemia were significantly associated with increased initial WBC count. Initial WBC count was significantly associated with hypoxia, pulmonary hemorrhage, CNS ischemia, CNS hemorrhage and death (Sung et al. 2012). Patients with the highest WBC counts (≥400 × 10⁹/L) were at greatest risk for early death and leukapheresis did not appear to impact early death: 1 of 16 (6.3 %) receiving leukapheresis experienced early death as compared to 3 of 73 (4.1 %) who were not leukapheresed (Sung et al. 2012). Overall, the induction death rate was 3.9 % for WBC ≥100 × 10⁹/L vs. 1.3 % for WBC <100 × 10⁹/L. In a comprehensive review of retrospective AML studies in both pediatric and adult patients with HL, Oberoi et al. (2014) reported an overall early death rate of 20.1 % which was not significantly influenced by the use of leukapheresis. Hemorrhage and bleeding (75 %) was the most common cause of early death followed by leukostasis (9.1 %); APL patients constituted only 2–5.5 % of all studied patients and pediatric data were not separately analyzed (Oberoi et al. 2014).

Inaba et al. (2008) studied newly diagnosed AML patients at St. Jude Children’s Research Hospital and found an 18.3 % incidence of HL; in the most recent period, the early death rate was 2.8 % as compared to a previous time period with an early death rate of 22.9 %. Complications secondary to HL were similar in both time periods and significantly correlated with median WBC count: neurologic (WBC 221 × 10⁹/L vs. 158 × 10⁹/L), pulmonary (240 × 10⁹/L vs. 155 × 10⁹/L), and renal (WBC 275 × 10⁹/L vs. 159 × 10⁹/L). Leukoreduction was utilized in the later time period at the discretion of the treating physician; early death occurred in 1 of 20 patients leukoreduced and 0 of 16 not leukoreduced although those leukoreduced had a significantly higher admission WBC count (206 × 10⁹/L vs. 116 × 10⁹/L). Metabolic complications were uncommon although more prevalent in those with FAB M4/M5 subtypes. Twelve of the 17 deaths occurred due to hemorrhage (gastrointestinal, pulmonary or intracranial).

Earlier studies have reported a much higher death rate, likely due in part to less intensive supportive care measures (Wald et al. 1982; Bunin and Pui 1985; Creutzig et al. 2004; Inaba et al. 2008). Although significant data are lacking, one study in adult AML M5 patients with HL and pulmonary symptoms has shown potential benefit with the addition of dexamethasone to improve lung function in the acute period (Azoulay et al. 2012). The utilization of leukapheresis in AML is controversial and discussed further in Sect. 6.7.

HL is an uncommon presenting feature in APL although pediatric patients have an increased incidence of microgranular APL (M3v) with concomitant HL (Rovelli et al. 1992; Guglielmi et al. 1998; de Botton et al. 2004). HL can more commonly be seen as part of APL differentiation syndrome (previously called retinoic acid syndrome) after commencement of either all-trans retinoic acid (ATRA) or arsenic trioxide (ATO) therapy (Vahdat et al. 1994; Camacho et al. 2000; Levy et al. 2008; Zhang et al. 2008; Sanz et al. 2009; Zhou et al. 2010). In some studies, increasing leukocytosis (WBC >10 × 10⁹/L) was a risk factor for the development of the clinical findings of APL differentiation syndrome such as unexplained fever, weight gain, respiratory distress, pulmonary infiltrates, pleural effusions or pericarditis (Frankel et al. 1992; Vahdat et al. 1994; Camacho et al. 2000). HL either at diagnosis or after initiation of differentiating therapy is a risk factor for early death, especially from ICH secondary to disseminated intravascular coagulation (DIC) (Rovelli et al. 1992; Roberts et al. 2000; Zhang et al. 2008; Zhou et al. 2010).

Due to the risk of early death from DIC and hemorrhage, especially in those with the highest WBC counts, it is vital to initiate supportive care and induction therapy with ATRA and/or ATO emergently (Sanz et al. 2005, 2009; Tallman and Altman 2009). In lieu of randomized data, expert guidelines recommend treatment of coagulopathy with fresh frozen plasma, fibrinogen (fibrinogen concentrate or cryoprecipitate), and platelet transfusion to maintain fibrinogen >150 mg/dL and platelets >50 × 10⁹/L at the minimum in those with HL, with frequent (i.e., every 6–8 h) monitoring and correction (Sanz et al. 2005, 2009; Tallman and Altman 2009). Studies on the complementary use of antifibrinolytics such as tranexamic acid have not shown benefit (Sanz et al. 2009). Diagnostic lumbar puncture and placement of a central venous catheter should be avoided until the coagulopathy has resolved (Sanz et al. 2005, 2009; Tallman and Altman 2009). Initiation of treatment with a differentiating agent should start prior to genetic confirmation of diagnosis in those cases with sufficient clinical suspicion (Sanz et al. 2005, 2009; Tallman and Altman 2009). The clinician should be cognizant of the risk for HL development and concomitant DIC after initiation of either ATRA or ATO.

In the pediatric North American INT0129 APL trial, hydroxyurea was initiated at a dose of 1 g/m² and ATRA held if the WBC rose to >30 × 10⁹/L during ATRA therapy until the WBC count was <10 × 10⁹/L (Gregory et al. 2009). In a Chinese study of 19 children with APL who received single-agent ATO, all developed an increase in WBC count with induction ATO with 5 having WBC >100 × 10⁹/L (Zhou et al. 2010). The two children with the highest WBC counts, 178 × 10⁹/L and 252 × 10⁹/L after ATO initiation, both died from ICH. Zhou et al. (2010) initiated hydroxyurea for all WBC >20 × 10⁹/L while also decreasing the ATO dose or even holding it in patients with severe leukocytosis. The benefit of holding ATO therapy with HL is unclear (Levy et al. 2008; Sanz et al. 2009). Oral corticosteroids, which are utilized as treatment for APL differentiation syndrome, have also been suggested as a prophylactic agent with WBC >5–50 × 10⁹/L or with the initiation of induction therapy in all patients to prevent the development of subsequent HL as well as severe pulmonary or CNS symptoms (Wiley and Firkin 1995; Sanz et al. 2005, 2009; Tallman and Altman 2009) Randomized data are lacking. The use of leukapheresis in APL with HL is controversial and discussed in Sect. 6.7.

HL is reported to be more common in pediatric patients with CML as compared to adults (Rowe and Lichtman 1984; Millot et al. 2005). Even with the high WBC counts at presentation, prevalence of leukostasis secondary to pediatric CML is thought to be uncommon although it was seen frequently in the small review by Rowe and Lichtman (1984). Symptoms of leukostasis in CML patients can include neurologic complaints such as papilledema, cranial nerve defects, and tinnitus, respiratory complaints, and priapism (Rowe and Lichtman 1984). Priapism has been noted in pediatric ALL and CML (Castagnetti et al. 2008; Vaitkevičienė et al. 2013). Management of HL in CML prior to the initiation of tyrosine kinase inhibitor therapy can be accomplished with hydroxyurea (Schwartz and Canellos 1975). Management of pediatric patients with signs and symptoms of leukostasis lacks an evidence basis in the literature although sources recommend utilization of chemotherapy in addition to leukapheresis (Rowe and Lichtman 1984; Castagnetti et al. 2008). For low-flow (ischemic) priapism in particular, adult guidelines recommend chemotherapy, leukapheresis, and urologic therapy with therapeutic aspiration and intracavernous sympathomimetics if aspiration alone is not successful (Montague et al. 2003; Rogers et al. 2012). Castagnetti et al. (2008) suggest patients can be managed without invasive urologic procedures utilizing chemotherapy and leukapheresis alone; however, their small cohort required a long period of time to recover from priapism and half did not receive leukapheresis. In addition they recommend the use of anticoagulation with low-molecular-weight heparin (especially with concomitant thrombocytosis) although significant evidence on utility is lacking. On long-term follow-up, none of their patient cohort had developed clinical evidence of erectile dysfunction.

The general management of tumor lysis syndrome (TLS) is discussed in Chap.​ 3. Patients with HL are at increased risk of both laboratory and clinical TLS, especially those with ALL. Truong et al. (2007) showed that WBC >20 × 10⁹/L was an independent risk factor for the development of TLS in pediatric ALL. Montesinos et al. (2008) similarly reported that WBC >25 × 10⁹/L was an independent risk factor for TLS in adult AML patients, but it is unclear if TLS is a significant issue in pediatric AML even with HL (Inaba et al. 2008; Sung et al. 2012). Prevention through the utilization of hyperhydration, urine alkalinization, and allopurinol are all proven methods to reduce the risk of metabolic complications in pediatric patients with HL and ALL (Maurer et al. 1988; Lascari 1991). Maurer et al. (1988) showed no benefit in preventing metabolic derangement through the use of low-dose prednisone prior to initiation of induction chemotherapy in pediatric ALL patients with WBC >200 × 10⁹/L.

The most recent expert TLS guidelines list both ALL and AML with concomitant HL as high-risk for the development of TLS and recommend prophylactic use of rasburicase in addition to aggressive hydration to prevent hyperuricemia (Cairo et al. 2010). These recommendations concurred with the Italian consensus guidelines for ALL, although in the Italian opinion TLS was rare in AML even with HL and therefore should not be considered high-risk (Tosi et al. 2008). Cairo et al. (2010) recommend one dose of rasburicase at a dose of 0.1–0.2 mg/kg prior to the initiation of therapy, while Tosi et al. (2008) recommend continuing daily therapy at a dose of 0.2 mg/kg/day for 3–5 days. Additionally, the concomitant use of allopurinol (due to the very rarely reported precipitation of uric acid precursor xanthine) and alkalinization (due to potential risk of calcium phosphate precipitation) are not recommended (Coiffier et al. 2008; Tosi et al. 2008; Howard et al. 2011; Vaitkevičienė et al. 2013). Rasburicase is contraindicated in patients with glucose-6-phosphate dehydrogenase (G6PD) deficiency and methemoglobinemia (Tosi et al. 2008; Cairo et al. 2010; Howard et al. 2011; Vaitkevičienė et al. 2013).

The most recent recommendations by the American Society for Apheresis (ASFA) suggest leukapheresis as first-line therapy for patients with leukostasis secondary to HL due to the potential to impact early death, although long-term outcome is unaffected (Schwartz et al. 2013). However, the potential to impact early death through leukapheresis was not shown in a meta-analysis of AML studies by Oberoi et al. (2014). The role of apheresis as leukostasis prophylaxis with HL is not established but may be considered (Schwartz et al. 2013). Per the ASFA guidelines, the utilization of leukapheresis is listed as a strong recommendation with moderate-quality evidence for treatment of leukostasis symptoms and should be considered as a weak recommendation for leukostasis prophylaxis in higher-risk AML patients (e.g., M4/M5 subtypes, rapidly rising blast count) with WBC >100 × 10⁹/L and ALL patients with WBC >400 × 10⁹/L (Schwartz et al. 2013). Per the ASFA guidelines, leukapheresis should not be utilized solely for the prevention or treatment of TLS in patients with HL (Schwartz et al. 2013). These recommendations are unchanged from the previous 2007 and 2010 ASFA guidelines (Szczepiorkowski et al. 2007, 2010). No specific contraindication to the use of leukapheresis in APL patients is included in these guidelines although it is mentioned as a relative contraindication in other guidelines due to the theoretical risk of worsening DIC and the increasing risk of ICH with lysis of leukemic promyelocytes (Blum and Porcu 2007; Szczepiorkowski et al. 2007; Sanz et al. 2009; Szczepiorkowski et al. 2010; Zuckerman et al. 2012; Kim and Sloan 2013; Schwartz et al. 2013). Data to support this contraindication are limited to one small study in which a majority of patients undergoing leukapheresis for HL had an adverse event that was not temporally related to the leukapheresis procedure (Vahdat et al. 1994; Tallman and Altman 2009). Strauss et al. (1985) successfully performed exchange transfusion on a 2-year-old child with APL and a presenting WBC count of 617 × 10⁹/L. Zuckerman et al. (2012) recommend leukapheresis in symptomatic adult AML patients with WBC >50 × 10⁹/L and in symptomatic ALL and CML adult patients with WBC >150 × 10⁹/L. Additionally, although stating a lack of evidence, they recommend leukapheresis in asymptomatic adult AML patients with WBC >100 × 10⁹/L to prevent leukostasis and asymptomatic adult ALL patients with WBC >300 × 10⁹/L to prevent TLS.