- Email: [email protected]
Blood transfusion in children with sickle cell disease undergoing tonsillectomy
International Journal of Pediatric Otorhinolaryngology 103 (2017) 117–120
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Blood transfusion in children with sickle cell disease undergoing tonsillectomy☆
MARK
Carlyn M. Atwooda,∗, Sharon H. Gnagia, Ronald J. Teufel IIb, Shaun A. Nguyena, David R. Whitea a b
Department of Otolaryngology Head & Neck Surgery, Medical University of South Carolina, Charleston, SC, USA Department of Pediatrics, Medical University of South Carolina, Charleston, SC, USA
A R T I C L E I N F O
A B S T R A C T
Keywords: Tonsillectomy Sickle cell Pediatrics
Introduction: Tonsillectomy is the second most common surgery in children with sickle cell disease. These children are at an increased risk of perioperative complications due to vaso-occlusive events. Although controversial, preoperative blood transfusions are sometimes given in an effort to prevent such complications. The purpose of this study is to analyze trends in the use of blood transfusion for management of children with sickle cell disease (SCD) undergoing tonsillectomy in a national database. Methods: Patients in the 1997–2012 KID with a primary procedure matching the ICD-9 procedure code for tonsillectomy (28.2–28.3) and diagnosis code for SCD (282.60–282.69) were examined. Patients were split into groups by blood transfusion status and compared across variables including complication rate, length of stay (LOS), and hospital charges. Statistical analysis included chi-square test for trend, Mann-Whitney U test, and independent t-test. Results: 1133 patients with SCD underwent tonsillectomy. There was a strong positive correlation between increasing chronologic year and the proportion of patients receiving blood transfusions, 47 (30.1%) in 1997 to 78 (42.5%) in 2012 (r = 0.94, p = 0.005). During this period, there was no significant change in the rate of complications (r = −0.1, p = 0.87). Overall, patients receiving blood transfusion had a longer mean LOS (3.1 ± 2.4 days vs. 2.5 ± 2.2 days, p < 0.005) and higher mean charge ($17,318 ± 13,191 vs. $13,532 ± 12,124, p < 0.005) compared to patients who did not receive blood transfusion. The rate of complications in the transfusion group, 18 of 352(5.1%), was not significantly different (p = 0.48) from the group without transfusion, 40 of 626 (6.4%). Conclusions: From 1997 to 2012, there was a significant increase in the proportion of patients with SCD receiving perioperative blood transfusions for tonsillectomy. While the frequency of transfusion rose, those who received a transfusion had similar complication rates with increased charges and length of hospital stays compared to those who did not receive a transfusion.
1. Introduction There are an estimated 100,000 people with sickle cell disease (SCD) in the US. Children account for nearly one third of this population and are more likely to undergo surgery than their peers without SCD [1,2]. Children with SCD often have tonsillar hypertrophy and/or recurrent tonsillitis necessitating the need for surgery and making adenotonsillectomy one of the most commonly performed procedures in children with SCD, second only to cholecystectomy [3]. Due to the defective hemoglobin in SCD (HbS), patients with SCD are at risk of developing vaso-occlusive events. The most common of these are acute pain crises, which can be precipitated by factors such as cold
temperature, infection, dehydration, physical exertion, anxiety, and hormonal changes associated with menses [4–6]. The added physiologic stressors of surgical trauma and general anesthesia complicate the situation. In addition to standard surgical complications such as hemorrhage and post-operative infection, patients with SCD are at an increased risk of vaso-occlusive complications during surgery. Insufficient oxygenation and circulatory stasis lead to HbS polymerization, which can cause vaso-occlusive crises, ischemic infarcts and acute chest syndrome. Patients with SCD who undergo tonsillectomy have been reported to have surgical complication rates of 21–32% [7,8]. Preoperative blood transfusions are commonly given to patients with SCD in an effort to decrease intraoperative and postoperative
☆ ∗
This was presented as an oral presentation at the Triological Society Combined Otolaryngology Spring Meeting in San Diego, California on April 28th, 2017. Corresponding author. 34 Mayfair Lane, Greenville, SC 29609, USA. E-mail address: [email protected] (C.M. Atwood).
http://dx.doi.org/10.1016/j.ijporl.2017.10.013 Received 6 August 2017; Received in revised form 28 September 2017; Accepted 5 October 2017 Available online 10 October 2017 0165-5876/ © 2017 Elsevier B.V. All rights reserved.
International Journal of Pediatric Otorhinolaryngology 103 (2017) 117–120
C.M. Atwood et al.
values and proportions were calculated using methods that take into account the complex sampling methods used in the KID. Unless otherwise specified, all data is weighted using HCUP supplied weight for generating national estimates. Values were considered significant at the level of p < 0.05.
complications by decreasing the percentage of HbS RBCs, improving anemia, and suppressing hematopoiesis [7]. A conservative approach consists of RBC transfusion to reach a hemoglobin concentration ≥10 g/dL, while a more aggressive approach incorporates exchange transfusion to lower the HbS to less than 30% [9]. The risks associated with any non-autogenic blood transfusion include acute and delayed transfusion reactions and potential transmission of blood-borne disease, thus the benefits to patients must outweigh the risks. Due to the potential harm to patients, clinical trials evaluating the safety and efficacy of perioperative management of children with SCD pose ethical issues and are very difficult to perform. The purpose of this study is to look at trends in the perioperative management of children with SCD who undergo tonsillectomy by using a large, national, de-identified, pediatric inpatient database from the years of 1997–2012.
3. Results 3.1. Overall population There were a total of 69,687 weighted admissions with the primary procedure of tonsillectomy with or without adenoidectomy (T ± A) in the 1997–2012 KID. Of these, 1133 were children with a diagnosis of SCD. Due to the non-normal distribution of the data, a Mann-Whitney U test was performed to assess if there were differences between the age at time of surgery between those with and without SCD. The age distributions of the two populations were not similar on visual inspection. Children with SCD were significantly older (7 years, IQR:5, 11) compared to the group without SCD (4 years, IQR:2, 8; p < 0.001). The median length of stay (LOS) was longer in the group with SCD (2 days, IQR:2,3) compared to the group without SCD (1 day, IQR:1,2) when analyzed using the Mann-Whitney U test (p < 0.001). Demographic information is listed in Table 1.
2. Methods 2.1. Data source The Kids' Inpatient Database (KID), part of the Healthcare Cost and Utilization Project (HCUP), is the largest national pediatric inpatient database in the US. The KID is released every 3 years with the most recent edition containing data from the year 2012. The database samples 80% of all pediatric discharges and 10% of all in-hospital births from all community, non-rehabilitation hospitals in participating states. Since its start in 1997, it has grown from 22 states to 44 states reporting data in 2012. The database includes information on the hospital course in the form International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9 CM) Diagnosis (DC) and Procedure (PC) and clinical classification software codes (CCS). Additionally, the database contains demographic information including patient age at time of admission, sex, race, and primary expected payer. Patient age at the time of admission is reported in whole number years, ranging from 0 to 18 in the 1997 KID and 0 to 20 in all editions thereafter. All data in the KID is de-identified, therefore this study did not require approval from the Office of Research Integrity at the Medical University of South Carolina. In compliance with the HCUP data use agreement, no values less than or equal to ten were reported in this study in order to protect patient anonymity.
3.2. Sickle cell disease In total, 399 (35.2%) of the children with SCD admitted for T ± A received a blood transfusion during the hospital course. In each KID year, exchange transfusion occurred in fewer than eleven admissions. Due to the small sample size of the exchange transfusion cohort, all RBC transfusions, including exchange transfusion, were grouped together for analysis. In 1997, 47 of the 155 admissions with SCD (30.1%) received blood transfusion during their hospital stay. In 2012, that number increased to 78 out of 184 admissions (42.5%). Chi-square analysis for trend shows a significant positive linear trend for the proportion of patients with SCD receiving blood transfusion over time from 1997 to 2012 (p = 0.004). The change in proportion of patients with SCD receiving a blood transfusion during their hospital stay for tonsillectomy is displayed in Fig. 1. The median age was equal in both the transfusion and no transfusion groups (7 years, IQR:5, 11) with no significant difference in the age distribution between the two groups (p = 0.86, Mann-Whitney U test). Tonsillar hypertrophy (ICD-9 CM 474.10–474.12) was the most common principal diagnosis (N = 825), accounting for 72.8% of all admissions. Chronic tonsillitis (ICD-9 CM 474.00–474.02) was the principal diagnosis for 14.8% of admissions (N = 168). There was no significant difference in principal diagnosis between the group that received blood transfusion and those who did not (Chi-square, p = 0.18). The mean hospital charges per year were consistently higher in the group that received blood transfusion compared to cohort that did not receive transfusion, although this difference did not reach significance until the year 2006 (Table 2). The overall median LOS (1997–2012) was the same in both groups, but the distribution of blood transfusion group was longer (2 days;
2.2. Data analysis The 1997, 2000, 2003, 2006, 2009, and 2012 KID were queried for admissions with tonsillectomy with or without adenoidectomy (ICD-9 CM PC 28.2–28.3) as the primary procedure, and containing a diagnosis of sickle cell disease (ICD-9 CM DC 292.60–292.69). Admissions which included blood transfusion (ICD-9 PC 99.00–99.04) were identified and tracked across all editions of the KID and compared to the admission that did not include blood transfusion. The ICD-9 CCS code for complication of surgical procedure or medical care (238) and the ICD-9 CM diagnosis code for acute chest syndrome (517.3) were used as markers for complications. CCS data is not available in the 1997 edition of the KID, therefore complication data is from the 2000–2012 KID only. The principal diagnosis was used as the indication for surgery (tonsillar hypertrophy, tonsillitis, or other). 2.3. Statistics
Table 1 Sickle cell status.
Statistical analysis was performed using Statistical Package for the Social Sciences (SPSS, Version 23; IBM SPSS Inc., Chicago IL). Ordinal data is reported as median and interquartile range. The Mann-Whitney U test was performed to determine differences in age of admission, length of stay, and total charges between groups. Chi-square test for homogeneity was used to compare proportions across populations. Chisquare for trend was used to determine if the proportion of patients undergoing blood transfusion changed significantly over time. All
N Median age (years) Median LOS LOS: Length of Stay (days). IQR: Interquartile Range.
118
SCD
No SCD
p
1133 7 (IQR:5, 11) 2 (IQR:2, 3)
68,363 4 (IQR:2, 8) 1 (IQR:1, 2)
< 0.001 < 0.001
International Journal of Pediatric Otorhinolaryngology 103 (2017) 117–120
C.M. Atwood et al.
Table 4 Complications.
Acute Chest Syndrome Medical/Surgical Complicationa
Blood Transfusion
No Blood Transfusion
P value
11 (2.76%) 18 (5.11%)
11 (1.50%) 40 (6.39%)
0.141 0.417
a Complications of surgical procedure or medical care are based on CCC code (238) and not available for the 1997 data set.
years with sleep disordered breathing are an exception. Due to increased risk of airway compromise, they are often kept inpatient overnight following tonsillectomy [10]. This young, otherwise healthy cohort may contribute to the lower mean age of the general population compared to those with SCD.
Fig. 1. Blood transfusions over time.
Table 2 Average total hospital charges (US dollars).
5. Complications
Year
Blood transfusion
No blood transfusion
p
1997 2000 2003 2006 2009 2012
8268 ± 5624 9195 ± 5423 14,213 ± 9998 18,174 ± 14,707 21,850 ± 15,436 22,010 ± 10,283
7620 ± 7743 8530 ± 5744 12,070 ± 14,716 15,644 ± 16,997 17,087 ± 9604 17,416 ± 9875
0.629 0.34 0.08 0.021 0.001 0.003
In sickle cell disease (SCD), affected red blood cells sickle in low oxygen tension environments leading to painful vaso-occlusive events. The severity of sickling is determined by the pH, temperature, circulatory status, and oxygen tension of the blood. Defective red blood cells are destroyed by the body resulting in a chronic hemolytic anemia. Complications seen in arise secondary to vaso-occlusive events as well as chronic anemia [11]. The perioperative environment poses many potential triggers for a vaso-occlusive event to occur. During an operation, the many aspects of homeostasis must be balanced while surgical trauma induces added stress on the body. Common postoperative complications in the SCD population include pain crises, hypoxia, fevers, acute chest syndrome, and pneumonia [8]. Intravenous hydration, thermoregulation, adequate oxygenation, and preoperative blood transfusions are all part of the management of patients aimed at preventing sickle cell crises from occurring [4]. There has been controversy in medical literature over the best transfusion regimen to decrease complications in the SCD population undergoing surgery. Several studies advocate the use of simple blood transfusions aimed at increasing hemoglobin levels to ≥10 g/dL over more aggressive regimens that incorporate exchange transfusions to decrease HbS to ≤30%. This recommendation is based on the lack of improvement in overall morbidity and the increase in transfusion related complications associated with aggressive transfusion [9,12–14] This contrasts with the findings by Halvorson et al. that support a more aggressive transfusion regimen based on an association of increased complications with HbS > 40% [15]. In a recently updated Cochrane review, there was no difference in all-cause mortality between people who received aggressive transfusions compared to those managed with a conservative regimen [16]. Despite the controversy in the literature regarding the best regimen, there has been an overall shift in favor of preoperative blood transfusion in the SCD population. Howard et al. conducted a randomized trial comparing outcomes in patients with SCD undergoing low to medium risk surgeries. One group received preoperative blood transfusion while the other did not. The rate of clinically relevant complications did not differ significantly between groups. However, there were more significant adverse events in the group that did not receive preoperative blood transfusion (p = 0.002) [7]. In particular, there was a significant proportion of patients in the group that did not receive preoperative blood transfusion that developed acute chest syndrome, leading to the early termination of the study. Despite low-quality evidence, the 2014 Expert Panel Report recommends that adults and children with sickle cell anemia receive RBCs transfusion to bring the hemoglobin level to 10 g/dL prior to undergoing surgery with general anesthesia [17]. This study demonstrates a statistically significant increase in the
Table 3 Median length of stay (Days). Year
Blood Transfusion
No Blood Transfusion
p
1997 2000 2003 2006 2009 2012
2.2 (IQR: 2, 4) 2.75 (IQR: 2, 4) 2 (IQR: 2, 3) 2 (IQR: 2, 3) 2 (IQR: 2, 3) 2 (IQR: 2, 3)
2 2 2 2 2 2
0.079 < 0.0005 0.002 0.05 < 0.0005 0.013
(IQR: (IQR: (IQR: (IQR: (IQR: (IQR:
2, 1, 1, 2, 1, 2,
4) 3) 3) 3) 2.4) 2)
IQR: Interquartile Range.
IQR:2, 3) compared to the group that did not receive a blood transfusion (2 days; IQR: 1,2; Mann-Whitney U test, p < 0.001). When LOS was analyzed by year, the mean rank was greater in the group that received blood transfusion compared to the group that did not for all years of the KID; this difference was statistically significant from 2000 to 2012 (Table 3). The overall complication rate, as measured by the CCS code for Complication of a Surgical Procedure or Medical Care, did not change significantly from 2000 (7.53%) to 2012 (7.03%; Cochran-Armitage test for trend, p = 0.504). The overall incidence of acute chest syndrome increased over the study period, from 0% in 1997 to 1.6% of patients in 2012 (N < 11) but did not reach statistical significance (p = 0.056). There was no significant difference in the rate of acute chest syndrome (p = 0.14) or in the aforementioned CCS (p = 0.42) between the groups based on transfusion status (Table 4). The occurrences of splenic sequestration, transfusion reactions, and hemorrhage complicating a procedure were too low to report.
4. Discussion Children with SCD were significantly older at the time of admission compared to their peers without SCD. The population with SCD also had a significantly longer length of stay (LOS). Children who are otherwise healthy often undergo tonsillectomy as an outpatient procedure and are not captured by the KID. Children under the age of three 119
International Journal of Pediatric Otorhinolaryngology 103 (2017) 117–120
C.M. Atwood et al.
for tonsillectomy. Blood transfusion was associated with an increase in hospital charges and length of stay without a significant change in perioperative complications. The increase costs, time spent in the hospital, and risk associated with transfusion may not result in a significant change in complications in children with sickle cell disease undergoing tonsillectomy, adding to the controversy surrounding this practice.
percentage of patients with SCD receiving blood transfusions during tonsillectomy admission from 1997 to 2012. This increase is likely driven by the previously mentioned increase in support for preoperative blood transfusions in the SCD population for prevention of perioperative complications. However, in spite of the increase in transfusions over the study period, there was no change the incidence of complications.
Conflict of interest and financial disclosures 6. Length of stay The authors do not have relevant conflicts of interest or financial disclosures to report.
When a preoperative transfusion protocol is implemented, it often requires the patient to be admitted the day prior to surgery in order for the transfusion to occur, thus, increasing the length of stay (LOS). This is consistent with our findings of a statistically significant increase in the LOS for admissions including a transfusion compared to those who did not receive a transfusion. This finding differs from Howard et al. who found no significant difference LOS between groups based on preoperative transfusion status [7].
References [1] O.S. Platt, D.J. Brambilla, W.F. Rosse, et al., Mortality in sickle cell disease. Life expectancy and risk factors for early death, N. Engl. J. Med. 330 (1994) 1639–1644. [2] K.L. Hassell, Population estimates of sickle cell disease in the U.S, Am. J. Prev. Med. 38 (2010) S512–S521. [3] O. Hyder, M. Yaster, B.T. Bateman, P.G. Firth, Surgical procedures and outcomes among children with sickle cell disease, Anesth. Analg. 117 (2013) 1192–1196. [4] D.S., J. Buck, Surgery in sickle cell disease, Hematol. Oncol. Clin. N. Am. 19 (2005) 897–902. [5] W.C. Yoong, S.M. Tuck, Menstrual pattern in women with sickle cell anaemia and its association with sickling crises, J. Obstet. Gynecol. 22 (2002) 399–401. [6] A.C. Stanley, J.M. Christian, Sickle cell disease and perioperative considerations: review and retrospective report, J. Oral Maxillofac. Surg. 71 (2013) 1027–1033. [7] J. Howard, M. Malfroy, C. Llewelyn, L. Choo, R. Hodge, T. Johnson, S. Purohit, D.C. Rees, L. Tillyer, I. Walker, K. Fijnvandraat, M. Kirby-Allen, E. Spackman, S.C. Davies, L.M. Williamson, The Transfusion Alternatives Preoperatively in Sickle Cell Disease (TAPS) study: a randomised, controlled, multicentre clinical trial, Lancet 381 (2013) 930–938. [8] R.L. Duke, J.P. Scott, J.A. Panepinto, V.A. Flanary, Perioperative management of sickle cell disease children undergoing adenotonsillectomy, Otolaryngol. Head. Neck Surg. 134 (2006) 370–373. [9] Y.A. Wali, H.A. Okbi, R.A. Abri, A comparison of two transfusion regimens in the perioperative management of children with sickle cell disease undergoing adenotonsillectomy, Pediatr. Hematol. Oncol. 20 (2009) 7–13. [10] R.F. Baugh, S.M. Archer, R.B. Mitchell, et al., Clinical practice guideline: tonsillectomy in children, Otolaryngol. Head Neck Surg. 144 (2011) S1–S30. [11] C.B. Derkay, G, G.J. Milmoe, K.M. Grundfast, Adenotonsillectomy in children with sickle cell disease, South. Med. J. 84 (1991) 205–208. [12] M.W. Koshy, SJ, S.T. Miller, L.A. Sleeper, E. Vichinsky, A.K. Brown, Y. Khakoo, T.R. Kinneythe Cooperative Study of Sicle Cell Disease, Surgery and anesthesia in sickle cell disease, Blood 86 (1995). [13] E.H. Vichinsky, CM, L. Neumar, A.N. Earles, D. Black, M. Koshy, C. Pegelowthe Perioperative Transfusion in Sickle Cell Disease Study Group, A comparison of conservative and aggressive transfusion regimens in the perioperative management of sickle cell disease, N. Eng. J. Med. 333 (1995) 206–213. [14] P.P. Waldron, C, L. Neumayr, C. Haberkern, A. Earles, R. Wesman, VichinskyE and the Preoperative Transfusions in Sickle Cell Disease Study Group, Tonsillectomy, adenoidectomy, and myringotomy in sickle cell disease: perioperative morbidity, J. Pediatr. Hematol. Oncol. 29 (1999) 129–135. [15] D.M. Halvorson, V, K. McKie, P.E. Ashmore, E.S. Porubsky, Sickle cell disease and tonsillectomy preoperative management and postoperative complications, Arch. Otolaryngol. Head. Neck Surg. 123 (1997) 689–692. [16] L.J. Estcourt, P.M. Fortin, M. Trivella, S. Hopewell, Preoperative blood transfusions for sickle cell disease, Cochrane Database Syst. Rev. 4 (2016) CD003149. [17] Evidence-Based Management of Sickle Cell Disease: Expert Panel Report, US Department of Health and Human Services, 2014.
7. Hospital charges In our study, the mean hospital charges per year were greater in the transfusion group compared to the non-transfused group. This finding differs from Howard et al. who found that a lower mean cost in the preoperative transfusion group (p = 0.399). Since the group that received transfusions in this study had a longer LOS, it is not surprising that the additional hospital stay would also drive up costs. 8. Limitations The main limitation of this study is attributed to the nature of the data source. The KID, a large, de-identified national database, may contain coding errors or inconsistencies that could lead to incorrect inclusion or exclusion of patients. Information within the KID is derived from hospital billing codes rather than medical records, thus, we are unable to determine the chronologic order of events or the indication for transfusion. Therefore, we cannot conclude if blood transfusions occurred prior to, during, or after tonsillectomy. Further investigation into the efficacy of preoperative blood transfusion in the sickle cell population is warranted. Despite its limitations, this study demonstrates a previously undocumented shift in the management of children with sickle cell disease undergoing tonsillectomy. In addition to shedding light on shifting practices in the US, this study comments on the effects of transfusions, such as increased costs and hospital stay. 9. Conclusion From 1997 to 2012, there was a significant increase in the proportion of patients with SCD receiving perioperative blood transfusions
120
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Blood transfusion in children with sickle cell disease undergoing tonsillectomy☆
MARK
Carlyn M. Atwooda,∗, Sharon H. Gnagia, Ronald J. Teufel IIb, Shaun A. Nguyena, David R. Whitea a b
Department of Otolaryngology Head & Neck Surgery, Medical University of South Carolina, Charleston, SC, USA Department of Pediatrics, Medical University of South Carolina, Charleston, SC, USA
A R T I C L E I N F O
A B S T R A C T
Keywords: Tonsillectomy Sickle cell Pediatrics
Introduction: Tonsillectomy is the second most common surgery in children with sickle cell disease. These children are at an increased risk of perioperative complications due to vaso-occlusive events. Although controversial, preoperative blood transfusions are sometimes given in an effort to prevent such complications. The purpose of this study is to analyze trends in the use of blood transfusion for management of children with sickle cell disease (SCD) undergoing tonsillectomy in a national database. Methods: Patients in the 1997–2012 KID with a primary procedure matching the ICD-9 procedure code for tonsillectomy (28.2–28.3) and diagnosis code for SCD (282.60–282.69) were examined. Patients were split into groups by blood transfusion status and compared across variables including complication rate, length of stay (LOS), and hospital charges. Statistical analysis included chi-square test for trend, Mann-Whitney U test, and independent t-test. Results: 1133 patients with SCD underwent tonsillectomy. There was a strong positive correlation between increasing chronologic year and the proportion of patients receiving blood transfusions, 47 (30.1%) in 1997 to 78 (42.5%) in 2012 (r = 0.94, p = 0.005). During this period, there was no significant change in the rate of complications (r = −0.1, p = 0.87). Overall, patients receiving blood transfusion had a longer mean LOS (3.1 ± 2.4 days vs. 2.5 ± 2.2 days, p < 0.005) and higher mean charge ($17,318 ± 13,191 vs. $13,532 ± 12,124, p < 0.005) compared to patients who did not receive blood transfusion. The rate of complications in the transfusion group, 18 of 352(5.1%), was not significantly different (p = 0.48) from the group without transfusion, 40 of 626 (6.4%). Conclusions: From 1997 to 2012, there was a significant increase in the proportion of patients with SCD receiving perioperative blood transfusions for tonsillectomy. While the frequency of transfusion rose, those who received a transfusion had similar complication rates with increased charges and length of hospital stays compared to those who did not receive a transfusion.
1. Introduction There are an estimated 100,000 people with sickle cell disease (SCD) in the US. Children account for nearly one third of this population and are more likely to undergo surgery than their peers without SCD [1,2]. Children with SCD often have tonsillar hypertrophy and/or recurrent tonsillitis necessitating the need for surgery and making adenotonsillectomy one of the most commonly performed procedures in children with SCD, second only to cholecystectomy [3]. Due to the defective hemoglobin in SCD (HbS), patients with SCD are at risk of developing vaso-occlusive events. The most common of these are acute pain crises, which can be precipitated by factors such as cold
temperature, infection, dehydration, physical exertion, anxiety, and hormonal changes associated with menses [4–6]. The added physiologic stressors of surgical trauma and general anesthesia complicate the situation. In addition to standard surgical complications such as hemorrhage and post-operative infection, patients with SCD are at an increased risk of vaso-occlusive complications during surgery. Insufficient oxygenation and circulatory stasis lead to HbS polymerization, which can cause vaso-occlusive crises, ischemic infarcts and acute chest syndrome. Patients with SCD who undergo tonsillectomy have been reported to have surgical complication rates of 21–32% [7,8]. Preoperative blood transfusions are commonly given to patients with SCD in an effort to decrease intraoperative and postoperative
☆ ∗
This was presented as an oral presentation at the Triological Society Combined Otolaryngology Spring Meeting in San Diego, California on April 28th, 2017. Corresponding author. 34 Mayfair Lane, Greenville, SC 29609, USA. E-mail address: [email protected] (C.M. Atwood).
http://dx.doi.org/10.1016/j.ijporl.2017.10.013 Received 6 August 2017; Received in revised form 28 September 2017; Accepted 5 October 2017 Available online 10 October 2017 0165-5876/ © 2017 Elsevier B.V. All rights reserved.
International Journal of Pediatric Otorhinolaryngology 103 (2017) 117–120
C.M. Atwood et al.
values and proportions were calculated using methods that take into account the complex sampling methods used in the KID. Unless otherwise specified, all data is weighted using HCUP supplied weight for generating national estimates. Values were considered significant at the level of p < 0.05.
complications by decreasing the percentage of HbS RBCs, improving anemia, and suppressing hematopoiesis [7]. A conservative approach consists of RBC transfusion to reach a hemoglobin concentration ≥10 g/dL, while a more aggressive approach incorporates exchange transfusion to lower the HbS to less than 30% [9]. The risks associated with any non-autogenic blood transfusion include acute and delayed transfusion reactions and potential transmission of blood-borne disease, thus the benefits to patients must outweigh the risks. Due to the potential harm to patients, clinical trials evaluating the safety and efficacy of perioperative management of children with SCD pose ethical issues and are very difficult to perform. The purpose of this study is to look at trends in the perioperative management of children with SCD who undergo tonsillectomy by using a large, national, de-identified, pediatric inpatient database from the years of 1997–2012.
3. Results 3.1. Overall population There were a total of 69,687 weighted admissions with the primary procedure of tonsillectomy with or without adenoidectomy (T ± A) in the 1997–2012 KID. Of these, 1133 were children with a diagnosis of SCD. Due to the non-normal distribution of the data, a Mann-Whitney U test was performed to assess if there were differences between the age at time of surgery between those with and without SCD. The age distributions of the two populations were not similar on visual inspection. Children with SCD were significantly older (7 years, IQR:5, 11) compared to the group without SCD (4 years, IQR:2, 8; p < 0.001). The median length of stay (LOS) was longer in the group with SCD (2 days, IQR:2,3) compared to the group without SCD (1 day, IQR:1,2) when analyzed using the Mann-Whitney U test (p < 0.001). Demographic information is listed in Table 1.
2. Methods 2.1. Data source The Kids' Inpatient Database (KID), part of the Healthcare Cost and Utilization Project (HCUP), is the largest national pediatric inpatient database in the US. The KID is released every 3 years with the most recent edition containing data from the year 2012. The database samples 80% of all pediatric discharges and 10% of all in-hospital births from all community, non-rehabilitation hospitals in participating states. Since its start in 1997, it has grown from 22 states to 44 states reporting data in 2012. The database includes information on the hospital course in the form International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9 CM) Diagnosis (DC) and Procedure (PC) and clinical classification software codes (CCS). Additionally, the database contains demographic information including patient age at time of admission, sex, race, and primary expected payer. Patient age at the time of admission is reported in whole number years, ranging from 0 to 18 in the 1997 KID and 0 to 20 in all editions thereafter. All data in the KID is de-identified, therefore this study did not require approval from the Office of Research Integrity at the Medical University of South Carolina. In compliance with the HCUP data use agreement, no values less than or equal to ten were reported in this study in order to protect patient anonymity.
3.2. Sickle cell disease In total, 399 (35.2%) of the children with SCD admitted for T ± A received a blood transfusion during the hospital course. In each KID year, exchange transfusion occurred in fewer than eleven admissions. Due to the small sample size of the exchange transfusion cohort, all RBC transfusions, including exchange transfusion, were grouped together for analysis. In 1997, 47 of the 155 admissions with SCD (30.1%) received blood transfusion during their hospital stay. In 2012, that number increased to 78 out of 184 admissions (42.5%). Chi-square analysis for trend shows a significant positive linear trend for the proportion of patients with SCD receiving blood transfusion over time from 1997 to 2012 (p = 0.004). The change in proportion of patients with SCD receiving a blood transfusion during their hospital stay for tonsillectomy is displayed in Fig. 1. The median age was equal in both the transfusion and no transfusion groups (7 years, IQR:5, 11) with no significant difference in the age distribution between the two groups (p = 0.86, Mann-Whitney U test). Tonsillar hypertrophy (ICD-9 CM 474.10–474.12) was the most common principal diagnosis (N = 825), accounting for 72.8% of all admissions. Chronic tonsillitis (ICD-9 CM 474.00–474.02) was the principal diagnosis for 14.8% of admissions (N = 168). There was no significant difference in principal diagnosis between the group that received blood transfusion and those who did not (Chi-square, p = 0.18). The mean hospital charges per year were consistently higher in the group that received blood transfusion compared to cohort that did not receive transfusion, although this difference did not reach significance until the year 2006 (Table 2). The overall median LOS (1997–2012) was the same in both groups, but the distribution of blood transfusion group was longer (2 days;
2.2. Data analysis The 1997, 2000, 2003, 2006, 2009, and 2012 KID were queried for admissions with tonsillectomy with or without adenoidectomy (ICD-9 CM PC 28.2–28.3) as the primary procedure, and containing a diagnosis of sickle cell disease (ICD-9 CM DC 292.60–292.69). Admissions which included blood transfusion (ICD-9 PC 99.00–99.04) were identified and tracked across all editions of the KID and compared to the admission that did not include blood transfusion. The ICD-9 CCS code for complication of surgical procedure or medical care (238) and the ICD-9 CM diagnosis code for acute chest syndrome (517.3) were used as markers for complications. CCS data is not available in the 1997 edition of the KID, therefore complication data is from the 2000–2012 KID only. The principal diagnosis was used as the indication for surgery (tonsillar hypertrophy, tonsillitis, or other). 2.3. Statistics
Table 1 Sickle cell status.
Statistical analysis was performed using Statistical Package for the Social Sciences (SPSS, Version 23; IBM SPSS Inc., Chicago IL). Ordinal data is reported as median and interquartile range. The Mann-Whitney U test was performed to determine differences in age of admission, length of stay, and total charges between groups. Chi-square test for homogeneity was used to compare proportions across populations. Chisquare for trend was used to determine if the proportion of patients undergoing blood transfusion changed significantly over time. All
N Median age (years) Median LOS LOS: Length of Stay (days). IQR: Interquartile Range.
118
SCD
No SCD
p
1133 7 (IQR:5, 11) 2 (IQR:2, 3)
68,363 4 (IQR:2, 8) 1 (IQR:1, 2)
< 0.001 < 0.001
International Journal of Pediatric Otorhinolaryngology 103 (2017) 117–120
C.M. Atwood et al.
Table 4 Complications.
Acute Chest Syndrome Medical/Surgical Complicationa
Blood Transfusion
No Blood Transfusion
P value
11 (2.76%) 18 (5.11%)
11 (1.50%) 40 (6.39%)
0.141 0.417
a Complications of surgical procedure or medical care are based on CCC code (238) and not available for the 1997 data set.
years with sleep disordered breathing are an exception. Due to increased risk of airway compromise, they are often kept inpatient overnight following tonsillectomy [10]. This young, otherwise healthy cohort may contribute to the lower mean age of the general population compared to those with SCD.
Fig. 1. Blood transfusions over time.
Table 2 Average total hospital charges (US dollars).
5. Complications
Year
Blood transfusion
No blood transfusion
p
1997 2000 2003 2006 2009 2012
8268 ± 5624 9195 ± 5423 14,213 ± 9998 18,174 ± 14,707 21,850 ± 15,436 22,010 ± 10,283
7620 ± 7743 8530 ± 5744 12,070 ± 14,716 15,644 ± 16,997 17,087 ± 9604 17,416 ± 9875
0.629 0.34 0.08 0.021 0.001 0.003
In sickle cell disease (SCD), affected red blood cells sickle in low oxygen tension environments leading to painful vaso-occlusive events. The severity of sickling is determined by the pH, temperature, circulatory status, and oxygen tension of the blood. Defective red blood cells are destroyed by the body resulting in a chronic hemolytic anemia. Complications seen in arise secondary to vaso-occlusive events as well as chronic anemia [11]. The perioperative environment poses many potential triggers for a vaso-occlusive event to occur. During an operation, the many aspects of homeostasis must be balanced while surgical trauma induces added stress on the body. Common postoperative complications in the SCD population include pain crises, hypoxia, fevers, acute chest syndrome, and pneumonia [8]. Intravenous hydration, thermoregulation, adequate oxygenation, and preoperative blood transfusions are all part of the management of patients aimed at preventing sickle cell crises from occurring [4]. There has been controversy in medical literature over the best transfusion regimen to decrease complications in the SCD population undergoing surgery. Several studies advocate the use of simple blood transfusions aimed at increasing hemoglobin levels to ≥10 g/dL over more aggressive regimens that incorporate exchange transfusions to decrease HbS to ≤30%. This recommendation is based on the lack of improvement in overall morbidity and the increase in transfusion related complications associated with aggressive transfusion [9,12–14] This contrasts with the findings by Halvorson et al. that support a more aggressive transfusion regimen based on an association of increased complications with HbS > 40% [15]. In a recently updated Cochrane review, there was no difference in all-cause mortality between people who received aggressive transfusions compared to those managed with a conservative regimen [16]. Despite the controversy in the literature regarding the best regimen, there has been an overall shift in favor of preoperative blood transfusion in the SCD population. Howard et al. conducted a randomized trial comparing outcomes in patients with SCD undergoing low to medium risk surgeries. One group received preoperative blood transfusion while the other did not. The rate of clinically relevant complications did not differ significantly between groups. However, there were more significant adverse events in the group that did not receive preoperative blood transfusion (p = 0.002) [7]. In particular, there was a significant proportion of patients in the group that did not receive preoperative blood transfusion that developed acute chest syndrome, leading to the early termination of the study. Despite low-quality evidence, the 2014 Expert Panel Report recommends that adults and children with sickle cell anemia receive RBCs transfusion to bring the hemoglobin level to 10 g/dL prior to undergoing surgery with general anesthesia [17]. This study demonstrates a statistically significant increase in the
Table 3 Median length of stay (Days). Year
Blood Transfusion
No Blood Transfusion
p
1997 2000 2003 2006 2009 2012
2.2 (IQR: 2, 4) 2.75 (IQR: 2, 4) 2 (IQR: 2, 3) 2 (IQR: 2, 3) 2 (IQR: 2, 3) 2 (IQR: 2, 3)
2 2 2 2 2 2
0.079 < 0.0005 0.002 0.05 < 0.0005 0.013
(IQR: (IQR: (IQR: (IQR: (IQR: (IQR:
2, 1, 1, 2, 1, 2,
4) 3) 3) 3) 2.4) 2)
IQR: Interquartile Range.
IQR:2, 3) compared to the group that did not receive a blood transfusion (2 days; IQR: 1,2; Mann-Whitney U test, p < 0.001). When LOS was analyzed by year, the mean rank was greater in the group that received blood transfusion compared to the group that did not for all years of the KID; this difference was statistically significant from 2000 to 2012 (Table 3). The overall complication rate, as measured by the CCS code for Complication of a Surgical Procedure or Medical Care, did not change significantly from 2000 (7.53%) to 2012 (7.03%; Cochran-Armitage test for trend, p = 0.504). The overall incidence of acute chest syndrome increased over the study period, from 0% in 1997 to 1.6% of patients in 2012 (N < 11) but did not reach statistical significance (p = 0.056). There was no significant difference in the rate of acute chest syndrome (p = 0.14) or in the aforementioned CCS (p = 0.42) between the groups based on transfusion status (Table 4). The occurrences of splenic sequestration, transfusion reactions, and hemorrhage complicating a procedure were too low to report.
4. Discussion Children with SCD were significantly older at the time of admission compared to their peers without SCD. The population with SCD also had a significantly longer length of stay (LOS). Children who are otherwise healthy often undergo tonsillectomy as an outpatient procedure and are not captured by the KID. Children under the age of three 119
International Journal of Pediatric Otorhinolaryngology 103 (2017) 117–120
C.M. Atwood et al.
for tonsillectomy. Blood transfusion was associated with an increase in hospital charges and length of stay without a significant change in perioperative complications. The increase costs, time spent in the hospital, and risk associated with transfusion may not result in a significant change in complications in children with sickle cell disease undergoing tonsillectomy, adding to the controversy surrounding this practice.
percentage of patients with SCD receiving blood transfusions during tonsillectomy admission from 1997 to 2012. This increase is likely driven by the previously mentioned increase in support for preoperative blood transfusions in the SCD population for prevention of perioperative complications. However, in spite of the increase in transfusions over the study period, there was no change the incidence of complications.
Conflict of interest and financial disclosures 6. Length of stay The authors do not have relevant conflicts of interest or financial disclosures to report.
When a preoperative transfusion protocol is implemented, it often requires the patient to be admitted the day prior to surgery in order for the transfusion to occur, thus, increasing the length of stay (LOS). This is consistent with our findings of a statistically significant increase in the LOS for admissions including a transfusion compared to those who did not receive a transfusion. This finding differs from Howard et al. who found no significant difference LOS between groups based on preoperative transfusion status [7].
References [1] O.S. Platt, D.J. Brambilla, W.F. Rosse, et al., Mortality in sickle cell disease. Life expectancy and risk factors for early death, N. Engl. J. Med. 330 (1994) 1639–1644. [2] K.L. Hassell, Population estimates of sickle cell disease in the U.S, Am. J. Prev. Med. 38 (2010) S512–S521. [3] O. Hyder, M. Yaster, B.T. Bateman, P.G. Firth, Surgical procedures and outcomes among children with sickle cell disease, Anesth. Analg. 117 (2013) 1192–1196. [4] D.S., J. Buck, Surgery in sickle cell disease, Hematol. Oncol. Clin. N. Am. 19 (2005) 897–902. [5] W.C. Yoong, S.M. Tuck, Menstrual pattern in women with sickle cell anaemia and its association with sickling crises, J. Obstet. Gynecol. 22 (2002) 399–401. [6] A.C. Stanley, J.M. Christian, Sickle cell disease and perioperative considerations: review and retrospective report, J. Oral Maxillofac. Surg. 71 (2013) 1027–1033. [7] J. Howard, M. Malfroy, C. Llewelyn, L. Choo, R. Hodge, T. Johnson, S. Purohit, D.C. Rees, L. Tillyer, I. Walker, K. Fijnvandraat, M. Kirby-Allen, E. Spackman, S.C. Davies, L.M. Williamson, The Transfusion Alternatives Preoperatively in Sickle Cell Disease (TAPS) study: a randomised, controlled, multicentre clinical trial, Lancet 381 (2013) 930–938. [8] R.L. Duke, J.P. Scott, J.A. Panepinto, V.A. Flanary, Perioperative management of sickle cell disease children undergoing adenotonsillectomy, Otolaryngol. Head. Neck Surg. 134 (2006) 370–373. [9] Y.A. Wali, H.A. Okbi, R.A. Abri, A comparison of two transfusion regimens in the perioperative management of children with sickle cell disease undergoing adenotonsillectomy, Pediatr. Hematol. Oncol. 20 (2009) 7–13. [10] R.F. Baugh, S.M. Archer, R.B. Mitchell, et al., Clinical practice guideline: tonsillectomy in children, Otolaryngol. Head Neck Surg. 144 (2011) S1–S30. [11] C.B. Derkay, G, G.J. Milmoe, K.M. Grundfast, Adenotonsillectomy in children with sickle cell disease, South. Med. J. 84 (1991) 205–208. [12] M.W. Koshy, SJ, S.T. Miller, L.A. Sleeper, E. Vichinsky, A.K. Brown, Y. Khakoo, T.R. Kinneythe Cooperative Study of Sicle Cell Disease, Surgery and anesthesia in sickle cell disease, Blood 86 (1995). [13] E.H. Vichinsky, CM, L. Neumar, A.N. Earles, D. Black, M. Koshy, C. Pegelowthe Perioperative Transfusion in Sickle Cell Disease Study Group, A comparison of conservative and aggressive transfusion regimens in the perioperative management of sickle cell disease, N. Eng. J. Med. 333 (1995) 206–213. [14] P.P. Waldron, C, L. Neumayr, C. Haberkern, A. Earles, R. Wesman, VichinskyE and the Preoperative Transfusions in Sickle Cell Disease Study Group, Tonsillectomy, adenoidectomy, and myringotomy in sickle cell disease: perioperative morbidity, J. Pediatr. Hematol. Oncol. 29 (1999) 129–135. [15] D.M. Halvorson, V, K. McKie, P.E. Ashmore, E.S. Porubsky, Sickle cell disease and tonsillectomy preoperative management and postoperative complications, Arch. Otolaryngol. Head. Neck Surg. 123 (1997) 689–692. [16] L.J. Estcourt, P.M. Fortin, M. Trivella, S. Hopewell, Preoperative blood transfusions for sickle cell disease, Cochrane Database Syst. Rev. 4 (2016) CD003149. [17] Evidence-Based Management of Sickle Cell Disease: Expert Panel Report, US Department of Health and Human Services, 2014.
7. Hospital charges In our study, the mean hospital charges per year were greater in the transfusion group compared to the non-transfused group. This finding differs from Howard et al. who found that a lower mean cost in the preoperative transfusion group (p = 0.399). Since the group that received transfusions in this study had a longer LOS, it is not surprising that the additional hospital stay would also drive up costs. 8. Limitations The main limitation of this study is attributed to the nature of the data source. The KID, a large, de-identified national database, may contain coding errors or inconsistencies that could lead to incorrect inclusion or exclusion of patients. Information within the KID is derived from hospital billing codes rather than medical records, thus, we are unable to determine the chronologic order of events or the indication for transfusion. Therefore, we cannot conclude if blood transfusions occurred prior to, during, or after tonsillectomy. Further investigation into the efficacy of preoperative blood transfusion in the sickle cell population is warranted. Despite its limitations, this study demonstrates a previously undocumented shift in the management of children with sickle cell disease undergoing tonsillectomy. In addition to shedding light on shifting practices in the US, this study comments on the effects of transfusions, such as increased costs and hospital stay. 9. Conclusion From 1997 to 2012, there was a significant increase in the proportion of patients with SCD receiving perioperative blood transfusions
120