- Email: [email protected]
Effect of honey on febrile neutropenia in children with acute lymphoblastic leukemia: A randomized crossover open-labeled study
Complementary Therapies in Medicine 25 (2016) 98–103
Contents lists available at ScienceDirect
Complementary Therapies in Medicine journal homepage: www.elsevierhealth.com/journals/ctim
Effect of honey on febrile neutropenia in children with acute lymphoblastic leukemia: A randomized crossover open-labeled study Mamdouh Abdulmaksoud Abdulrhman a,∗,1 , Ahmad Alsaeed Hamed a , Sahar Abdelmaksoud Mohamed b , Nouran Abdel Aal Hassanen a a b
Pediatric Department, Faculty of Medicine, Ain Shams University, Egypt Faculty of Medicine, El-Azhar University, Egypt
a r t i c l e
i n f o
Article history: Received 14 March 2015 Received in revised form 27 January 2016 Accepted 29 January 2016 Available online 2 February 2016 Keywords: Honey Febrile neutropenia Leukemia Chemotherapy
a b s t r a c t Objectives: Febrile neutropenia (FN) is a common adverse effect of chemotherapy. Current management of FN is expensive and may induce side effects. Honey, as a natural honeybee product, has antioxidant, antimicrobial, immunomodulator and anticancer effects. Additionally, honey is not expensive. The aim of this study is to test the effects of a 12-week honey consumption on children with acute lymphoblastic leukemia (ALL) particularly with regard to FN. Design: A randomized crossover clinical trial. Forty patients of both sexes, aged 2.5–10 years, were randomized into two equal groups [intervention to control (I/C) and control to intervention (C/I)]. Setting: Children Hospital of Ain Shams University-Cairo-Egypt. The dietary intervention consisted of honey in a dose of 2.5 g//kg body weight per dose twice weekly for 12 weeks. Main outcome measure: Febrile neutropenia in terms of frequency and duration of hospital admission. Results: The intervention resulted in a significant decrease of FN episodes, the number of patients admitted with FN and the duration of hospital stay. Also, honey consumption improved the levels of hemoglobin and did not produce any serious side effect. As a possible effect of honey withdrawal in the I/C group, the Hb%, the absolute neutrophil count and the platelet count decreased. Conclusion: Honey intervention in a group of children with ALL resulted in positive effects on FN and hematologic parameters. Further studies that include a larger number of patients are recommended to confirm that honey, has beneficial effects, as a complementary agent, in children with ALL. © 2016 Elsevier Ltd. All rights reserved.
1. Introduction Children with acute lymphoblastic leukemia (ALL) are at high risk of sepsis during the neutropenic phase following chemotherapy. In developing countries, sepsis is a major obstacle toward improving survival of affected children.1 Febrile neutropenia (FN) is defined as fever ≥38.3 ◦ C (101 ◦ F) or ≥38.0 ◦ C (100.4 ◦ F) for ≥1 h; and neutropenia is defined as a neutrophil count <500 cells/mm3 ; or <1000 cells/mm3 with a predicted decrease to <500 cells/mm3 .2 Despite advances in cancer treatment with chemotherapy and supportive care, FN remains a common
∗ Corresponding author. E-mail addresses: [email protected] (M.A. Abdulrhman), [email protected] (A.A. Hamed), [email protected] (S.A. Mohamed), [email protected] (N.A.A. Hassanen). 1 Present address: 4/11 Misr Lltaameer Bds, Sabel El Momineen St, El Sefarat, Madinat Nasr, Cairo, Egypt. http://dx.doi.org/10.1016/j.ctim.2016.01.009 0965-2299/© 2016 Elsevier Ltd. All rights reserved.
complication after chemotherapy.3 Although FN-related morbidity and mortality have been reduced, the currently available management strategies might induce side effects, including promoting microbial resistance against antibiotics, decreasing the patient’s quality of life, and increasing treatment costs.4,5 In ALL, the immune defects and the immunosuppressive effects of chemotherapy increase the risk for infections. The chemotherapy not only induces neutropenia, but also results in chemotactic and phagocytic defects.6,7 Based on the observations that honey, as a natural, inexpensive substance produced by honey bees, has antimicrobial,8–12 antioxidant,13–15 immunomodulator16 and anticancer effects,17–19 we tried in the present study to test the effects of honey on a group of children with ALL particularly with regard to FN episodes. Furthermore, honey can stimulate monocytes in cell cultures to produce tumor necrosis factor (TNF)-␣ and interleukin (IL)-1.20 These inflammatory mediators can stimulate tissues to produce Granulocyte Colony Stimulating Factor (G-CSF) which induces the production of neutrophils in chemotherapy-induced bone marrow suppression.21
M.A. Abdulrhman et al. / Complementary Therapies in Medicine 25 (2016) 98–103
2. Patients and methods 2.1. Patients This was a crossover randomized study that took place at the Children Hospital of Ain Shams University-Cairo-Egypt, from March 2011 to August 2013. Patients with ALL were recruited from the Outpatient Hematology-Oncology Clinic of the hospital. All patients >2 years of age with ALL, treated according to the Modified CCG 1991 protocol for standard- risk ALL and on maintenance therapy, were candidates for this study. All patients were receiving trimethoprim/sulfamethoxazole as a prophylactic antibiotic. Patients with diabetes mellitus (DM) and patients who had FN at the time enrollment were excluded from the study. 2.2. Study protocol A crossover design (two 12-week intervention periods) was used to measure honey effects. The subjects were randomized into either group A (control to intervention [C/I] group) or group B (intervention to control [I/C] group). Participants were randomly assigned following simple randomization procedure (computerized random numbers) to 1 of 2 groups with a 1:1 allocation ratio. Based on the results of the study of Zidan et al.22 which showed that honey intervention resulted in a decrease in the frequency of severe FN by 40%; a total sample size of 32 patients (16 per group) is required to have a 90% chance of detecting, as significant at the 5% level, a decrease in the frequency of FN by 40% during the intervention period. The subjects in the I/C group B took 2 ml (2.5 g) honey/kg body weight/dose twice weekly during the first 12-week period (period 1), whereas the subjects in the C/I group A did not receive honey as a control in the period 1. All parents were instructed not to give any type of bee honey to their children during the control period. The study was open labeled, and the dose of honey was empirical because there was no identification of a particular dose of honey in earlier clinical trials using honey in various diseases. Furthermore, dose-related toxicity to honey has not been previously reported, except probably for patients with DM. Following period 1, the protocol was exchanged for each group for the following 12-week period (period 2). Because laboratory tests for measurement of honey levels in blood or tissues are not yet available, each patient took the dose of honey under visual supervision of the researcher to ensure adherence. For this reason, the patients were seen in the clinic twice weekly during the intervention periods, whereas during the control periods they were seen once weekly. Each calculated dose of honey was dissolved in water with a ratio of 1:3, respectively, and then ingested by the patient before meal. Dissolving honey in water enhances the anti-microbial properties of honey,8,9 facilitates swallowing and helps adjusting the dose.
99
heat.23 Microscopic examination of samples from honey confirmed the presence of pollen grains, which were mainly of clover (Trifolium alexandrinum). The physicochemical characteristics of the honey used in the study are detailed in the Supplementary Table 1.24
2.4. Outcome measures The primary outcome measure was febrile neutropenia in terms of frequency, the number of patients admitted and the duration of hospital stay. The secondary outcome measures were hemoglobin (Hb) level, total leucocytic count (TLC), absolute neutrophil count (ANC) and platelet count (PLT). Blood count was performed for all patients on a weekly basis. All measures were analyzed in participants at baseline (0 week), the end of the 12th week (crossover) and the end of the 24th week (end point). All patients who developed FN during the study were admitted to the hospital and received an empirical combination of intravenous antibiotic therapy consisting of piperacillin (200 mg/kg/24 h, divided q 6 h) and amikacin (20 mg/kg q 24 h). Chemotherapy dose escalation was done when ANC exceeded 2000/mm3 for more than 6 weeks. All study procedures were in accordance with the ethical standards of the responsible institutional committee on human experimentation and with the Helsinki Declaration of 1975 (as revised in 1983). An informed consent was obtained from at least one parent of each child before enrollment, and the study was approved by the local Ethics Committee of the Pediatric Department of Ain Shams University Hospitals.
2.5. Statistical analysis The collected data were introduced to a PC using Statistical Package for Social Sciences (SPSS 15.0.1 for Windows; SPSS, Inc., 2001). Variables were expressed as mean (±SD) or median. Student’s t test was used for normal variables. The paired t-test was used to assess the statistical significance of the difference between two means measured twice for the same study group. Independent sample t-test was used to assess the statistical significance of the difference between two study group means. The Mann–Whitney U and Wilcoxon Signed Ranks tests were used for non-parametric data. Chi-square (2 ) test was used to measure the association between two qualitative variables. McNemar’s test was used for paired nominal data. A value of P < 0.05 was considered significant.
3. Results 2.3. Honey The honey used in this study was a raw, unprocessed (not heated or irradiated) clover honey collected from Al Mahala-Gharbia Governorate, Egypt. It was supplied directly from an apiary, and then kept in dark containers at room temperature for use in the study. Physicochemical analysis of the honey was done in the Chemical Analysis Laboratory of Honey Bee Products, Beekeeping Research Center, Plant Protection Research Institute, Agriculture Research Center, Giza, Egypt. This honey had a pH of 3.7; moisture content of 18.8%; electrical conductivity of 0.27 mS/cm; and a carbohydrate content of 78.4 g/100 g, with a fructose to glucose ratio of 1.2:0.8, respectively, and a non-reducing sugar content of 3.4 g/100 g. The Hydroxymethylfurfuraldehyde (HMF) content was 1.6 mg/kg. Values of HMF less than 15 mg/kg indicate fresh honey not exposed to
Fifty five patients were enrolled and assessed for eligibility. Eleven patients were excluded; 7 patients declined to participate, and 4 patients did not meet the inclusion criteria; one patient had DM and 3 had febrile neutropenia at baseline. Forty four eligible patients were thus randomized to either group A (C/I) or group B (I/C). One patient from group A lost follow-up, and 3 patients (one from group A and two from group B) discontinued the intervention because they developed undesirable effects following honey intake. Only 40 patients who completed the study protocol were included in the final statistical analysis (Fig. 1; flow chart). All enrolled patients were of low socioeconomic standard. Throughout the periods of the study no patient relapsed and no patient required blood transfusion, colony-stimulating factor (CSF) or central venous line.
100
M.A. Abdulrhman et al. / Complementary Therapies in Medicine 25 (2016) 98–103
Fig. 1. Flow chart.
Table 1 Baseline characteristics of studied groups. Characteristic
Group A (C/I)
Group B (I/C)
P
Age (yrs.) Male (%) Body weight (kg) Duration of disease (yr.) Median (IQR) Time interval (mo.) between start of maintenance therapy and onset of study Median (IQR) Hb% (g/dl) TLC (×103 /mm3 ) ANC (×103 /mm3 ) PLT (×103 /mm3 )
5.1 ± 2.5 50 19.6 ± 7.1 1.5 (0.75)
5.65 ± 2.24 50 19.6 ± 8.5 1.5 (2.25)
0.47†
15 (9)
15 (27)
0.89‡
10.75 ± 1.45 4.56 ± 1.94 2.05 ± 1.64 265.75 ± 78.24
11.5 ± 1.94 5.12 ± 1.39 2.29 ± 1.41 308.15 ± 50.06
0.17† 0.3† 0.64† 0.048†
0.99† 0.89‡
Hb—hemoglobin; TLC—total leucocytic count; ANC—absolute neutrophil count; PLT—platelet count. † Unpaired t-test. ‡ Mann–Whitney U-test.
3.1. Baseline characteristics of study groups The baseline characteristics of patients are shown in Table 1. The age of patients of both groups ranged from 2.5 to 10 years with a mean (SD) of 5.4 (2.4) years. There was no statistical significant difference between the two groups as regards the age, sex and body weight. Also, the duration of chemotherapy and the time interval between the start of maintenance therapy and the onset of the study did not differ significantly between groups. The hematologic parameters; Hb%, total leucocytic count (TLC) and absolute neutrophil count (ANC) did not differ significantly between the two groups. However, the platelet count (PLT) was slightly significantly lower in group A (p = 0.048). 3.2. Crossover comparison of periods 1 and 2 Table 2 compares the changes in the patient characteristics in periods 1 and 2. In group A (control to intervention group), period 1 [baseline vs. crossover (0 vs. 12th week)] represents the control period, whereas in group B (intervention to control group), period 1(baseline vs. crossover) represents the intervention period. In group A (C/I); a highly statistically significant decrease was found in the Hb% (p = 0.000) and a slight statistical significant decrease was found in PLT (p = 0.03), but the TLC and ANC did not show sig-
nificant differences at the end of the control period (crossover or 12th week) compared to baseline. However, in group B (I/C); no statistical significant difference observed in the Hb%, TLC, ANC and PLT at the end of the intervention (crossover) compared to baseline. On the other hand, period 2 [crossover vs. study end point (12th week vs. 24th week)] in group A (C/I) represents the intervention period, whereas in group B (I/C), it represents the control period. In group A (C/I); a statistical significant increase was observed in the Hb% (p = 0.000), ANC (p = 0.000) and PLT (P = 0.006) at the end of the intervention (study endpoint, or 24th week) compared to crossover. In group B (I/C); a highly statistically significant decrease was observed in the Hb% (p = 0.000), ANC (p = 0.000) and PLT (p = 0.000) at the end of the control period (study endpoint) compared to crossover. As a possible effect of the honey intervention, the ANC increased in group A, and it decreased as a possible effect of honey withdrawal in group B (Table 2). However, when the effects of the honey intervention on the number of the patients who required chemotherapy dose escalation as a result of the persistence of their ANC above 2000/mm3 for more than 6 weeks, were compared with the control in both groups, no statistically significant difference observed between the intervention and the control periods. In group A, one patient during the control period and also one patient in the intervention period
M.A. Abdulrhman et al. / Complementary Therapies in Medicine 25 (2016) 98–103
101
Table 2 Changes in the characteristic values of the study groups during periods 1 and 2. Characteristics
C/I
I/C
Control (period 1)
Intervention (period 1) Intervention (period 2)
Hb (g/dl) TLC (×103 /mm3 ) ANC (×103 /mm3 ) PLT (×103 /mm3 )
Baseline
12 weeks
10.75 ± 1.45 4.56 ± 1.94 2.05 ± 1.64 265.8 ± 78.2
8.12 4.73 1.42 214.4
± ± ± ±
0.49** 2.37 0.42 50.6*
Control (period 2) 24 weeks 10.46 5.19 1.97 269.0
± ± ± ±
0.76** 1.41 0.45** 57.8**
Baseline
12 weeks
11.5 ± 1.94 5.12 ± 1.39 2.29 ± 1.41 308.2 ± 50.1
11.3 5.67 2.61 280.1
± ± ± ±
1.23 1.64 1.15 41.2
24 weeks 8.57 4.69 1.34 225.2
± ± ± ±
1.14** 2.16 0.41** 41.1**
Values are means ± SD. Student’s paired t-test was used to compare the differences in the values in each study period (period 1, baseline vs. crossover [12 weeks]; period 2, crossover vs. study end point [24 weeks]) in each study group. Hb—hemoglobin; TLC—total leucocytic count; ANC—absolute neutrophil count; PLT—platelet count. * P < 0.05. ** P < 0.01.
required dose escalation, whereas in group B, five patients during the intervention and 4 patients during the control period required dose escalation, with no statistically significant difference between the intervention and control (p = 0.3).
3.3. Effects of honey on febrile neutropenia All patients with FN < 1000/mm3 were admitted to the hospital. The mean (SD) duration of hospital stay was 7.87 (3.32) days with a range from 3 to 15 days. The effects of honey on both FN < 500/mm3 and FN = 500–999//mm3 as regards the three measures; the frequency of FN, the number (%) of patients who developed FN and the sum of days of hospital stay, were analyzed in both groups (Table 3). In group A (C/I), these 3 measures did not show significant difference between the control and the intervention periods in the patients with FN < 500//mm3 , whereas in the patients with FN = 500–999//mm3 , the only significant difference was found in the number of these patients, which decreased significantly during the intervention period as compared with the control period. In group B (I/C), where the control period followed the intervention period, there was a significant increase of these three measures in the patients with FN < 500//mm3 during the control period, as a possible effect of honey withdrawal; whereas in the patients with FN = 500–999//mm3 , no significant difference observed between the intervention and the control periods. Comparing the two groups during period 1 i.e., comparing the control period of group A with the intervention period of group B, as regards the effects of honey on the 3 measures; the frequency of FN, the number (%) of patients who developed FN and the sum of days of hospital stay, found no significant difference in the patients with FN < 500/mm3 (p > 0.05). However, in the patients with FN = 500–999/mm3 , a statistically significant decrease of the FN episodes and the duration of hospital stay was observed during the intervention period of group B as compared with the control period of group A (p = 0.004 and 0.006, respectively), but the number of these patients did not show significant difference (p = 0.63). In both groups combined (40 patients), the incidence of FN < 500/mm3 and FN = 500–999/mm3 was 47.5 and 60%, respectively. As shown in Table 4 and Fig. 2, the effects of the honey intervention on the three FN measures and on the number of the patients in whom chemotherapy dose escalation was done, were compared with the control in both groups combined (40 patients). The only significant change observed was in the number of the patients who developed FN < 500/mm3 , which decreased during the intervention as compared with the control period. On the other hand, the FN episodes and the duration of hospital stay did not show significant difference. Also, the number of the patients in whom
Fig. 2. The effects of the honey Intervention (I) compared to Control (C) in both groups combined (40 patients).
chemotherapy dose escalation was done did not differ significantly between the intervention and the control. 3.4. Safety of honey No patient developed any serious adverse event as a result of the honey intervention. However, of the total 44 eligible patients, 10 (22.7%) patients developed undesirable effects in the form of abdominal pain, vomiting, fever, diarrhea, following honey intake. Three patients stopped the intervention as a result of these effects, whereas 7 continued the intervention. These undesirable effects are detailed in Supplementary Table 2. Of the 10 patients, 6 patients developed all these effects, whereas 4 patients developed only abdominal pain and vomiting. All these undesirable effects resolved spontaneously within a few minutes to a few hours without progression to any serious adverse event. The undesirable effects did not follow each dose of honey, but only some and not in a sequential manner. 4. Discussion The crossover design of this study enabled the effects of honey to be measured not only between different subjects in the same study period (period 1) but also in the same subjects in a crossover comparison of periods 1 and 2. In both groups the absolute neutrophil count (ANC) increased as a result of the honey intervention. This increase was significant in group A and non-significant in group B. Also, a significant decline in the ANC was found from the crossover to the study endpoint (period 2) as a possible effect of honey withdrawal in the I/C group B. It seems from these results that honey could result in a situation where oral chemotherapy is being increased, leading to a
102
M.A. Abdulrhman et al. / Complementary Therapies in Medicine 25 (2016) 98–103
Table 3 Febrile neutropenia (FN) among studied groups. FN
Group A (C/I)
No. = 20
Group B (I/C)
No. = 20
Control
Intervention
P
Intervention
Control
P
FN < 500/mm3 No. of episodes No. (%) of patients who developed FN Total No. of days of hospital admission
8 8 (40) 44
8 7 (35)§ 72
1† 0.3017‡ 0.75656†
2 2 (10) 12
11 10 (50)§ 104
0.034† 0.000‡ 0.0151†
FN = 500–999/mm3 No. of episodes No. (%)of patients who developed FN Total No. of days of hospital admission
19 14 (70)¶ 150
11 11 (55) 82
0.09692† 0.03689‡ 0.06148†
7 6 (30)§ 58
5 4 (20)§ 37
0.528695† 0.10035‡ 0.214975†
§ ¶ † ‡
One patient developed two episodes. Five patients developed two episodes. Wilcoxon Signed Ranks Test. McNemar’s test.
Table 4 Febrile neutropenia (FN) of both groups combined (40 patients). FN
Groups A and B No = 40 patients Control
Intervention
P
FN < 500/mm3 No. of episodes No. (%) of patients who developed FN Total No. of days of hospital admission
19 18 (45)§ 148
10 9 (22.5)§ 84
0.13104† 0.00004‡ 0.12602†
FN = 500–999/mm3 No. of episodes No. (%) of patients who developed FN Total No. of days of hospital admission
24 18 (45)¶ 187
18 17 (42.5)§ 140
0.29834† 0.71992‡ 0.29834†
No. (%) of patients in whom chemotherapy dose escalation was done
5 (12.5)
6 (15)
0.58388‡
§ ¶ † ‡
One patient developed two episodes. Six patients developed two episodes. Wilcoxon Signed Ranks Test. McNemar’s test.
vicious circle and possibly increased FN. However, we did not find a statistically significant difference between the control and the intervention periods of either group A or B as regards the number of patients who required chemotherapy dose escalation as a result of the persistence of their ANC above 2000/mm3 for more than 6 weeks. This means that the increased ANC as an effect of honey did not reach a level necessitating dose escalation. The possible positive effects of honey in reducing the frequency of FN, the number of patients admitted and the duration of hospital stay were evident in patients of group B with FN < 500/mm3 where there was a significant increase of these parameters as a possible effect of honey withdrawal. In the case of FN = 500–999/mm3 , a slightly significant reduction was only observed in the number of group A patients admitted as a possible result of the intervention. When we compared the previous measures between the subjects during period 1, i.e., the intervention period of group B with the control period of group A, significant positive effects were only observed in the patients with FN = 500–999/mm3 where the frequency of FN and the duration of hospital stay were significantly reduced as a possible effect of the intervention, but the number of patients admitted did not show significant difference. When we combined both groups together (total 40 patients) and we compared the control with the intervention period, we found that the positive effect of honey was only evident in the number of patients admitted with FN < 500/mm3 . The observed positive effects of honey on FN might be related to its antimicrobial,8–12 antioxidant13–15 and immunomodulator16 effects. Also, it might be hypothesized that honey, through increasing TNF-␣ and IL-120 might increase the levels of G-CSF. The incidence of febrile neutropenia observed in our study (> 40%) is high as compared with high income countries. The low
socioeconomic status of our patients with the prevalence of illiteracy and poverty probably contributes to this high rate. The possible positive effects of honey on hemoglobin (Hb) % and platelet count (PLT) were also evident during this study. Significant increases in Hb% and PLT were observed as a result of the intervention in group A. Conversely, significant reductions in Hb% and PLT were observed after honey withdrawal in the I/C group B. By performing a systematic search of PubMed using the terms honey, leukemia, cancer, chemotherapy, neutropenia, we only identified the study of Zidan et al.22 . In this study a type of honey called Life-Mel Honey (LMH) was found to be effective in reducing the incidence of anemia in 64% of the patients and decreasing the incidence of severe neutropenia (absolute neutrophils <500/mm3 ) by 40%. This study differed from our study in three main aspects: first, LMH is a honey that is produced by honey bees fed medicinal plants, whereas the honey used in our study was an Egyptian clover honey (T. alexandrinum L; Fam.: Leguminosae) that is produced by bees that feed primarily on the nectar of clover plants. Second, the study of Zidan et al. compared a group of patients before and after honey intake, whereas our study was a crossover study, which enabled a comparison of the effects of honey in two comparable groups of patients and in the same group of patients. Third, in the study of Zidan et al., the honey was given in a dose of 5 g/day for only 5 days, whereas in our study, honey was given in a larger dose; 2.5 g/kg twice weekly for 12 weeks for a total duration of 24 days. Despite these differences the possible positive effects of honey found in our study were nearly comparable with the results of Zidan et al. In our study, the incidence of severe neutropenia (<500/mm3 ) decreased from 47.5 to 25% as a result of the intervention in both groups combined (40 patients).
M.A. Abdulrhman et al. / Complementary Therapies in Medicine 25 (2016) 98–103
During the intervention periods, adverse effects in the form of abdominal pain, vomiting, diarrhea, fever, followed honey intake in 10 (22.7%) patients from both groups combined. These adverse effects may represent the misleading reactions to honey administration, which are benign reactions in the form of a transient aggravation of symptoms and signs of underlying illnesses.25 These reactions usually resolve spontaneously and thought to be a result of the immunomodulatory effects of honey.16,26 5. Study limitations The sample size is small. Follow up was limited to three months and the study was not blinded. A washout period was not calculated in this study because there is not yet available test to measure honey or its components in blood, and hence the half-life of honey concentration in the blood cannot be known. Therefore, as a limitation to this study, the possible carry-over effects of honey may yield statistical bias. Also a further chemical analysis of the honey used in the study to measure the phenolic content and other elements was not done. 6. Conclusion Honey intervention in a group of children with ALL may reduce the frequency of FN, the number of patients admitted with FN and the duration of their hospital stay. Further studies that include a larger number of patients are recommended to confirm that honey, has beneficial effects as a complementary agent, in children with ALL. Conflicts of interest None Contributors All authors contributed in designing the work, acquisition, analysis, and interpretation of data, drafting and revising the manuscript, and gave final approval of the version to be published. Acknowledgments All authors declare no support from any organization for the submitted work, no financial relationships with any organizations that might have an interest in the submitted work, and no other relationships or activities that could appear to have influenced the submitted work. The patients in the current study are among those followed up and admitted freely in the Children‘s hospital, Faculty of Medicine, Ain Shams University. We thank very much all patients and their parents for their participation in this study and their cooperation throughout the study periods. We also thank Ahmad M Sheded, lecturer of pediatrics, for statistical analysis. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ctim.2016.01. 009.
103
References 1. Yadav SP, Dua V, Sachdeva A. Sepsis is a major barrier to improving survival in childhood acute lymphoblastic leukemia in the developing world. J Pediatr Hematol Oncol. 2011;33(8):636. 2. Hughes WT, Armstrong D, Bodey GP. 2002 guidelines for the use of antimicrobial agents in neutropenic patients with cancer. Clin Infect Dis. 2002;34(6):730–751. 3. Sezgin G, Acipayam C, Ozkan A, Bayram I, Tanyeli A. Meropenem versus piperacillin-tazobactam as empiric therapy for febrile neutropenia in pediatric oncology patients. Asian Pac J Cancer Prev. 2014;15(11):4549–4553. 4. Ziino O, Tucci F, Rossi MR. Outpatient management of febrile neutropenia in children with cancer. Pediatr Rep. 2011;24:e8, 3 (1). 5. Castagnola E, Paola D, Giacchino R. Economic aspects of empiric antibiotic therapy for febrile neutropenia in children with cancer. Support Care Cancer. 1998;6(6):524–528. 6. Blay JY, Chauvin F, Le Cesne A. Early lymphopenia after cytotoxic chemotherapy as a risk factor for febrile neutropenia. J Clin Oncol. 1996;14:636. 7. Lockhart PB, Sonis ST. Relationship of oral complications to peripheral blood leukocyte and platelet counts in patients receiving cancer chemotherapy. Oral Surg Oral Med Oral Pathol. 1979;48:21. 8. Molan PC. The antibacterial activity of honey. 1. The nature of the antibacterial activity. Bee World. 1992;73:5–28. 9. Molan PC. The antibacterial activity of honey. 2. Variation in the potency of the antibacterial activity. Bee World. 1992;73:59–76. 10. Steinberg D, Kaine G, Gedalia I. Antibacterial effect of propolis and honey on oral bacteria. Am J Dent. 1996;9:236–239. 11. Molan PC. Honey as an antimicrobial agent. In: Mizrahi A, Lensky Y, eds. Bee Products: Properties, Applications and Apitherapy. London: Plenum; 1997:pp27. 12. Theunissen F, Grobler S, Gedalia I. The antifungal action of three South African honeys on Candida albicans. Apidologie. 2001;32:371–379. 13. Frankel S, Robinson GE, Berenbaum MR. Antioxidant capacity and correlated characteristics of 14 unifloral honeys. J Apic Res. 1998;37:27–31. 14. Gheldof N, Engeseth NJ. Antioxidant capacity of honeys from various floral sources based on the determination of oxygen radical absorbance capacity and inhibition of in vitro lipoprotein oxidation in human serum samples. J Agric Food Chem. 2002;50:3050–3055. 15. Gross HB, Polagruto JA, Zhu QY, et al. Effect of honey consumption on plasma antioxidant status in human subjects. Pap Am Chem Soc. 2004;227:U29. 16. Majtan J. Honey: an immunomodulator in wound healing. Wound Repair Regen. 2014;22(2):187–192. 17. Morales P, Haza AI. Antiproliferative and apoptotic effects of Spanish honeys. Pharmacogn Mag. 2013;9(35):231–237. 18. Abubakar MB, Abdullah WZ, Sulaiman SA, Suen AB. A review of molecular mechanisms of the anti-leukemic effects of phenolic compounds in honey. Int J Mol Sci. 2012;15:15054–15073, 13 (11). 19. Woo KJ, Jeong YJ, Park JW, Kwon TK. Chrysin-induced apoptosis is mediated through caspase activation and Akt inactivation in U937 leukemia cells. Biochem Biophys Res Commun. 2004;24:1215––1222, 325 (4). 20. Tonks AJ, Cooper RA, Jones KP, Blair S, Parton J, Tonks A. Honey stimulates inflammatory cytokine production from monocytes. Cytokine. 2003;7:242–247, 21 (5). 21. Bendalla LJ, Bradstockb KF. G-CSF: from granulopoietic stimulant to bone marrow stem cell mobilizing agent. Cytokine Growth Factor Rev. 2014;25(4):355–367. 22. Zidan J, Shetver L, Gershuny A, et al. Prevention of chemotherapy-induced ANCutropenia by special honey intake. Med Oncol. 2006;23(4):549–552. 23. Bogdanov S, Book of Honey. Honey Composition (Chapter 5). Bee Product Science. September, pp. 10. www.bee-hexagon.net; 2009 Accessed October 10.10.14. 24. Crane E. A Book of Honey. Oxford, U.K: Oxford University Press; 2016:1980. 25. Abdulrhman M. Lessons from honey. In: Paper Presented at the First International Conference on Api Phytotherapy, May 18–20. 2012. 26. Bogdanov S, Book of Honey. Honey for Nutrition and Health (Chapter 7). Bee Product Science. September 2009: pp. 10. www.bee-hexagon.net; 2014.
Contents lists available at ScienceDirect
Complementary Therapies in Medicine journal homepage: www.elsevierhealth.com/journals/ctim
Effect of honey on febrile neutropenia in children with acute lymphoblastic leukemia: A randomized crossover open-labeled study Mamdouh Abdulmaksoud Abdulrhman a,∗,1 , Ahmad Alsaeed Hamed a , Sahar Abdelmaksoud Mohamed b , Nouran Abdel Aal Hassanen a a b
Pediatric Department, Faculty of Medicine, Ain Shams University, Egypt Faculty of Medicine, El-Azhar University, Egypt
a r t i c l e
i n f o
Article history: Received 14 March 2015 Received in revised form 27 January 2016 Accepted 29 January 2016 Available online 2 February 2016 Keywords: Honey Febrile neutropenia Leukemia Chemotherapy
a b s t r a c t Objectives: Febrile neutropenia (FN) is a common adverse effect of chemotherapy. Current management of FN is expensive and may induce side effects. Honey, as a natural honeybee product, has antioxidant, antimicrobial, immunomodulator and anticancer effects. Additionally, honey is not expensive. The aim of this study is to test the effects of a 12-week honey consumption on children with acute lymphoblastic leukemia (ALL) particularly with regard to FN. Design: A randomized crossover clinical trial. Forty patients of both sexes, aged 2.5–10 years, were randomized into two equal groups [intervention to control (I/C) and control to intervention (C/I)]. Setting: Children Hospital of Ain Shams University-Cairo-Egypt. The dietary intervention consisted of honey in a dose of 2.5 g//kg body weight per dose twice weekly for 12 weeks. Main outcome measure: Febrile neutropenia in terms of frequency and duration of hospital admission. Results: The intervention resulted in a significant decrease of FN episodes, the number of patients admitted with FN and the duration of hospital stay. Also, honey consumption improved the levels of hemoglobin and did not produce any serious side effect. As a possible effect of honey withdrawal in the I/C group, the Hb%, the absolute neutrophil count and the platelet count decreased. Conclusion: Honey intervention in a group of children with ALL resulted in positive effects on FN and hematologic parameters. Further studies that include a larger number of patients are recommended to confirm that honey, has beneficial effects, as a complementary agent, in children with ALL. © 2016 Elsevier Ltd. All rights reserved.
1. Introduction Children with acute lymphoblastic leukemia (ALL) are at high risk of sepsis during the neutropenic phase following chemotherapy. In developing countries, sepsis is a major obstacle toward improving survival of affected children.1 Febrile neutropenia (FN) is defined as fever ≥38.3 ◦ C (101 ◦ F) or ≥38.0 ◦ C (100.4 ◦ F) for ≥1 h; and neutropenia is defined as a neutrophil count <500 cells/mm3 ; or <1000 cells/mm3 with a predicted decrease to <500 cells/mm3 .2 Despite advances in cancer treatment with chemotherapy and supportive care, FN remains a common
∗ Corresponding author. E-mail addresses: [email protected] (M.A. Abdulrhman), [email protected] (A.A. Hamed), [email protected] (S.A. Mohamed), [email protected] (N.A.A. Hassanen). 1 Present address: 4/11 Misr Lltaameer Bds, Sabel El Momineen St, El Sefarat, Madinat Nasr, Cairo, Egypt. http://dx.doi.org/10.1016/j.ctim.2016.01.009 0965-2299/© 2016 Elsevier Ltd. All rights reserved.
complication after chemotherapy.3 Although FN-related morbidity and mortality have been reduced, the currently available management strategies might induce side effects, including promoting microbial resistance against antibiotics, decreasing the patient’s quality of life, and increasing treatment costs.4,5 In ALL, the immune defects and the immunosuppressive effects of chemotherapy increase the risk for infections. The chemotherapy not only induces neutropenia, but also results in chemotactic and phagocytic defects.6,7 Based on the observations that honey, as a natural, inexpensive substance produced by honey bees, has antimicrobial,8–12 antioxidant,13–15 immunomodulator16 and anticancer effects,17–19 we tried in the present study to test the effects of honey on a group of children with ALL particularly with regard to FN episodes. Furthermore, honey can stimulate monocytes in cell cultures to produce tumor necrosis factor (TNF)-␣ and interleukin (IL)-1.20 These inflammatory mediators can stimulate tissues to produce Granulocyte Colony Stimulating Factor (G-CSF) which induces the production of neutrophils in chemotherapy-induced bone marrow suppression.21
M.A. Abdulrhman et al. / Complementary Therapies in Medicine 25 (2016) 98–103
2. Patients and methods 2.1. Patients This was a crossover randomized study that took place at the Children Hospital of Ain Shams University-Cairo-Egypt, from March 2011 to August 2013. Patients with ALL were recruited from the Outpatient Hematology-Oncology Clinic of the hospital. All patients >2 years of age with ALL, treated according to the Modified CCG 1991 protocol for standard- risk ALL and on maintenance therapy, were candidates for this study. All patients were receiving trimethoprim/sulfamethoxazole as a prophylactic antibiotic. Patients with diabetes mellitus (DM) and patients who had FN at the time enrollment were excluded from the study. 2.2. Study protocol A crossover design (two 12-week intervention periods) was used to measure honey effects. The subjects were randomized into either group A (control to intervention [C/I] group) or group B (intervention to control [I/C] group). Participants were randomly assigned following simple randomization procedure (computerized random numbers) to 1 of 2 groups with a 1:1 allocation ratio. Based on the results of the study of Zidan et al.22 which showed that honey intervention resulted in a decrease in the frequency of severe FN by 40%; a total sample size of 32 patients (16 per group) is required to have a 90% chance of detecting, as significant at the 5% level, a decrease in the frequency of FN by 40% during the intervention period. The subjects in the I/C group B took 2 ml (2.5 g) honey/kg body weight/dose twice weekly during the first 12-week period (period 1), whereas the subjects in the C/I group A did not receive honey as a control in the period 1. All parents were instructed not to give any type of bee honey to their children during the control period. The study was open labeled, and the dose of honey was empirical because there was no identification of a particular dose of honey in earlier clinical trials using honey in various diseases. Furthermore, dose-related toxicity to honey has not been previously reported, except probably for patients with DM. Following period 1, the protocol was exchanged for each group for the following 12-week period (period 2). Because laboratory tests for measurement of honey levels in blood or tissues are not yet available, each patient took the dose of honey under visual supervision of the researcher to ensure adherence. For this reason, the patients were seen in the clinic twice weekly during the intervention periods, whereas during the control periods they were seen once weekly. Each calculated dose of honey was dissolved in water with a ratio of 1:3, respectively, and then ingested by the patient before meal. Dissolving honey in water enhances the anti-microbial properties of honey,8,9 facilitates swallowing and helps adjusting the dose.
99
heat.23 Microscopic examination of samples from honey confirmed the presence of pollen grains, which were mainly of clover (Trifolium alexandrinum). The physicochemical characteristics of the honey used in the study are detailed in the Supplementary Table 1.24
2.4. Outcome measures The primary outcome measure was febrile neutropenia in terms of frequency, the number of patients admitted and the duration of hospital stay. The secondary outcome measures were hemoglobin (Hb) level, total leucocytic count (TLC), absolute neutrophil count (ANC) and platelet count (PLT). Blood count was performed for all patients on a weekly basis. All measures were analyzed in participants at baseline (0 week), the end of the 12th week (crossover) and the end of the 24th week (end point). All patients who developed FN during the study were admitted to the hospital and received an empirical combination of intravenous antibiotic therapy consisting of piperacillin (200 mg/kg/24 h, divided q 6 h) and amikacin (20 mg/kg q 24 h). Chemotherapy dose escalation was done when ANC exceeded 2000/mm3 for more than 6 weeks. All study procedures were in accordance with the ethical standards of the responsible institutional committee on human experimentation and with the Helsinki Declaration of 1975 (as revised in 1983). An informed consent was obtained from at least one parent of each child before enrollment, and the study was approved by the local Ethics Committee of the Pediatric Department of Ain Shams University Hospitals.
2.5. Statistical analysis The collected data were introduced to a PC using Statistical Package for Social Sciences (SPSS 15.0.1 for Windows; SPSS, Inc., 2001). Variables were expressed as mean (±SD) or median. Student’s t test was used for normal variables. The paired t-test was used to assess the statistical significance of the difference between two means measured twice for the same study group. Independent sample t-test was used to assess the statistical significance of the difference between two study group means. The Mann–Whitney U and Wilcoxon Signed Ranks tests were used for non-parametric data. Chi-square (2 ) test was used to measure the association between two qualitative variables. McNemar’s test was used for paired nominal data. A value of P < 0.05 was considered significant.
3. Results 2.3. Honey The honey used in this study was a raw, unprocessed (not heated or irradiated) clover honey collected from Al Mahala-Gharbia Governorate, Egypt. It was supplied directly from an apiary, and then kept in dark containers at room temperature for use in the study. Physicochemical analysis of the honey was done in the Chemical Analysis Laboratory of Honey Bee Products, Beekeeping Research Center, Plant Protection Research Institute, Agriculture Research Center, Giza, Egypt. This honey had a pH of 3.7; moisture content of 18.8%; electrical conductivity of 0.27 mS/cm; and a carbohydrate content of 78.4 g/100 g, with a fructose to glucose ratio of 1.2:0.8, respectively, and a non-reducing sugar content of 3.4 g/100 g. The Hydroxymethylfurfuraldehyde (HMF) content was 1.6 mg/kg. Values of HMF less than 15 mg/kg indicate fresh honey not exposed to
Fifty five patients were enrolled and assessed for eligibility. Eleven patients were excluded; 7 patients declined to participate, and 4 patients did not meet the inclusion criteria; one patient had DM and 3 had febrile neutropenia at baseline. Forty four eligible patients were thus randomized to either group A (C/I) or group B (I/C). One patient from group A lost follow-up, and 3 patients (one from group A and two from group B) discontinued the intervention because they developed undesirable effects following honey intake. Only 40 patients who completed the study protocol were included in the final statistical analysis (Fig. 1; flow chart). All enrolled patients were of low socioeconomic standard. Throughout the periods of the study no patient relapsed and no patient required blood transfusion, colony-stimulating factor (CSF) or central venous line.
100
M.A. Abdulrhman et al. / Complementary Therapies in Medicine 25 (2016) 98–103
Fig. 1. Flow chart.
Table 1 Baseline characteristics of studied groups. Characteristic
Group A (C/I)
Group B (I/C)
P
Age (yrs.) Male (%) Body weight (kg) Duration of disease (yr.) Median (IQR) Time interval (mo.) between start of maintenance therapy and onset of study Median (IQR) Hb% (g/dl) TLC (×103 /mm3 ) ANC (×103 /mm3 ) PLT (×103 /mm3 )
5.1 ± 2.5 50 19.6 ± 7.1 1.5 (0.75)
5.65 ± 2.24 50 19.6 ± 8.5 1.5 (2.25)
0.47†
15 (9)
15 (27)
0.89‡
10.75 ± 1.45 4.56 ± 1.94 2.05 ± 1.64 265.75 ± 78.24
11.5 ± 1.94 5.12 ± 1.39 2.29 ± 1.41 308.15 ± 50.06
0.17† 0.3† 0.64† 0.048†
0.99† 0.89‡
Hb—hemoglobin; TLC—total leucocytic count; ANC—absolute neutrophil count; PLT—platelet count. † Unpaired t-test. ‡ Mann–Whitney U-test.
3.1. Baseline characteristics of study groups The baseline characteristics of patients are shown in Table 1. The age of patients of both groups ranged from 2.5 to 10 years with a mean (SD) of 5.4 (2.4) years. There was no statistical significant difference between the two groups as regards the age, sex and body weight. Also, the duration of chemotherapy and the time interval between the start of maintenance therapy and the onset of the study did not differ significantly between groups. The hematologic parameters; Hb%, total leucocytic count (TLC) and absolute neutrophil count (ANC) did not differ significantly between the two groups. However, the platelet count (PLT) was slightly significantly lower in group A (p = 0.048). 3.2. Crossover comparison of periods 1 and 2 Table 2 compares the changes in the patient characteristics in periods 1 and 2. In group A (control to intervention group), period 1 [baseline vs. crossover (0 vs. 12th week)] represents the control period, whereas in group B (intervention to control group), period 1(baseline vs. crossover) represents the intervention period. In group A (C/I); a highly statistically significant decrease was found in the Hb% (p = 0.000) and a slight statistical significant decrease was found in PLT (p = 0.03), but the TLC and ANC did not show sig-
nificant differences at the end of the control period (crossover or 12th week) compared to baseline. However, in group B (I/C); no statistical significant difference observed in the Hb%, TLC, ANC and PLT at the end of the intervention (crossover) compared to baseline. On the other hand, period 2 [crossover vs. study end point (12th week vs. 24th week)] in group A (C/I) represents the intervention period, whereas in group B (I/C), it represents the control period. In group A (C/I); a statistical significant increase was observed in the Hb% (p = 0.000), ANC (p = 0.000) and PLT (P = 0.006) at the end of the intervention (study endpoint, or 24th week) compared to crossover. In group B (I/C); a highly statistically significant decrease was observed in the Hb% (p = 0.000), ANC (p = 0.000) and PLT (p = 0.000) at the end of the control period (study endpoint) compared to crossover. As a possible effect of the honey intervention, the ANC increased in group A, and it decreased as a possible effect of honey withdrawal in group B (Table 2). However, when the effects of the honey intervention on the number of the patients who required chemotherapy dose escalation as a result of the persistence of their ANC above 2000/mm3 for more than 6 weeks, were compared with the control in both groups, no statistically significant difference observed between the intervention and the control periods. In group A, one patient during the control period and also one patient in the intervention period
M.A. Abdulrhman et al. / Complementary Therapies in Medicine 25 (2016) 98–103
101
Table 2 Changes in the characteristic values of the study groups during periods 1 and 2. Characteristics
C/I
I/C
Control (period 1)
Intervention (period 1) Intervention (period 2)
Hb (g/dl) TLC (×103 /mm3 ) ANC (×103 /mm3 ) PLT (×103 /mm3 )
Baseline
12 weeks
10.75 ± 1.45 4.56 ± 1.94 2.05 ± 1.64 265.8 ± 78.2
8.12 4.73 1.42 214.4
± ± ± ±
0.49** 2.37 0.42 50.6*
Control (period 2) 24 weeks 10.46 5.19 1.97 269.0
± ± ± ±
0.76** 1.41 0.45** 57.8**
Baseline
12 weeks
11.5 ± 1.94 5.12 ± 1.39 2.29 ± 1.41 308.2 ± 50.1
11.3 5.67 2.61 280.1
± ± ± ±
1.23 1.64 1.15 41.2
24 weeks 8.57 4.69 1.34 225.2
± ± ± ±
1.14** 2.16 0.41** 41.1**
Values are means ± SD. Student’s paired t-test was used to compare the differences in the values in each study period (period 1, baseline vs. crossover [12 weeks]; period 2, crossover vs. study end point [24 weeks]) in each study group. Hb—hemoglobin; TLC—total leucocytic count; ANC—absolute neutrophil count; PLT—platelet count. * P < 0.05. ** P < 0.01.
required dose escalation, whereas in group B, five patients during the intervention and 4 patients during the control period required dose escalation, with no statistically significant difference between the intervention and control (p = 0.3).
3.3. Effects of honey on febrile neutropenia All patients with FN < 1000/mm3 were admitted to the hospital. The mean (SD) duration of hospital stay was 7.87 (3.32) days with a range from 3 to 15 days. The effects of honey on both FN < 500/mm3 and FN = 500–999//mm3 as regards the three measures; the frequency of FN, the number (%) of patients who developed FN and the sum of days of hospital stay, were analyzed in both groups (Table 3). In group A (C/I), these 3 measures did not show significant difference between the control and the intervention periods in the patients with FN < 500//mm3 , whereas in the patients with FN = 500–999//mm3 , the only significant difference was found in the number of these patients, which decreased significantly during the intervention period as compared with the control period. In group B (I/C), where the control period followed the intervention period, there was a significant increase of these three measures in the patients with FN < 500//mm3 during the control period, as a possible effect of honey withdrawal; whereas in the patients with FN = 500–999//mm3 , no significant difference observed between the intervention and the control periods. Comparing the two groups during period 1 i.e., comparing the control period of group A with the intervention period of group B, as regards the effects of honey on the 3 measures; the frequency of FN, the number (%) of patients who developed FN and the sum of days of hospital stay, found no significant difference in the patients with FN < 500/mm3 (p > 0.05). However, in the patients with FN = 500–999/mm3 , a statistically significant decrease of the FN episodes and the duration of hospital stay was observed during the intervention period of group B as compared with the control period of group A (p = 0.004 and 0.006, respectively), but the number of these patients did not show significant difference (p = 0.63). In both groups combined (40 patients), the incidence of FN < 500/mm3 and FN = 500–999/mm3 was 47.5 and 60%, respectively. As shown in Table 4 and Fig. 2, the effects of the honey intervention on the three FN measures and on the number of the patients in whom chemotherapy dose escalation was done, were compared with the control in both groups combined (40 patients). The only significant change observed was in the number of the patients who developed FN < 500/mm3 , which decreased during the intervention as compared with the control period. On the other hand, the FN episodes and the duration of hospital stay did not show significant difference. Also, the number of the patients in whom
Fig. 2. The effects of the honey Intervention (I) compared to Control (C) in both groups combined (40 patients).
chemotherapy dose escalation was done did not differ significantly between the intervention and the control. 3.4. Safety of honey No patient developed any serious adverse event as a result of the honey intervention. However, of the total 44 eligible patients, 10 (22.7%) patients developed undesirable effects in the form of abdominal pain, vomiting, fever, diarrhea, following honey intake. Three patients stopped the intervention as a result of these effects, whereas 7 continued the intervention. These undesirable effects are detailed in Supplementary Table 2. Of the 10 patients, 6 patients developed all these effects, whereas 4 patients developed only abdominal pain and vomiting. All these undesirable effects resolved spontaneously within a few minutes to a few hours without progression to any serious adverse event. The undesirable effects did not follow each dose of honey, but only some and not in a sequential manner. 4. Discussion The crossover design of this study enabled the effects of honey to be measured not only between different subjects in the same study period (period 1) but also in the same subjects in a crossover comparison of periods 1 and 2. In both groups the absolute neutrophil count (ANC) increased as a result of the honey intervention. This increase was significant in group A and non-significant in group B. Also, a significant decline in the ANC was found from the crossover to the study endpoint (period 2) as a possible effect of honey withdrawal in the I/C group B. It seems from these results that honey could result in a situation where oral chemotherapy is being increased, leading to a
102
M.A. Abdulrhman et al. / Complementary Therapies in Medicine 25 (2016) 98–103
Table 3 Febrile neutropenia (FN) among studied groups. FN
Group A (C/I)
No. = 20
Group B (I/C)
No. = 20
Control
Intervention
P
Intervention
Control
P
FN < 500/mm3 No. of episodes No. (%) of patients who developed FN Total No. of days of hospital admission
8 8 (40) 44
8 7 (35)§ 72
1† 0.3017‡ 0.75656†
2 2 (10) 12
11 10 (50)§ 104
0.034† 0.000‡ 0.0151†
FN = 500–999/mm3 No. of episodes No. (%)of patients who developed FN Total No. of days of hospital admission
19 14 (70)¶ 150
11 11 (55) 82
0.09692† 0.03689‡ 0.06148†
7 6 (30)§ 58
5 4 (20)§ 37
0.528695† 0.10035‡ 0.214975†
§ ¶ † ‡
One patient developed two episodes. Five patients developed two episodes. Wilcoxon Signed Ranks Test. McNemar’s test.
Table 4 Febrile neutropenia (FN) of both groups combined (40 patients). FN
Groups A and B No = 40 patients Control
Intervention
P
FN < 500/mm3 No. of episodes No. (%) of patients who developed FN Total No. of days of hospital admission
19 18 (45)§ 148
10 9 (22.5)§ 84
0.13104† 0.00004‡ 0.12602†
FN = 500–999/mm3 No. of episodes No. (%) of patients who developed FN Total No. of days of hospital admission
24 18 (45)¶ 187
18 17 (42.5)§ 140
0.29834† 0.71992‡ 0.29834†
No. (%) of patients in whom chemotherapy dose escalation was done
5 (12.5)
6 (15)
0.58388‡
§ ¶ † ‡
One patient developed two episodes. Six patients developed two episodes. Wilcoxon Signed Ranks Test. McNemar’s test.
vicious circle and possibly increased FN. However, we did not find a statistically significant difference between the control and the intervention periods of either group A or B as regards the number of patients who required chemotherapy dose escalation as a result of the persistence of their ANC above 2000/mm3 for more than 6 weeks. This means that the increased ANC as an effect of honey did not reach a level necessitating dose escalation. The possible positive effects of honey in reducing the frequency of FN, the number of patients admitted and the duration of hospital stay were evident in patients of group B with FN < 500/mm3 where there was a significant increase of these parameters as a possible effect of honey withdrawal. In the case of FN = 500–999/mm3 , a slightly significant reduction was only observed in the number of group A patients admitted as a possible result of the intervention. When we compared the previous measures between the subjects during period 1, i.e., the intervention period of group B with the control period of group A, significant positive effects were only observed in the patients with FN = 500–999/mm3 where the frequency of FN and the duration of hospital stay were significantly reduced as a possible effect of the intervention, but the number of patients admitted did not show significant difference. When we combined both groups together (total 40 patients) and we compared the control with the intervention period, we found that the positive effect of honey was only evident in the number of patients admitted with FN < 500/mm3 . The observed positive effects of honey on FN might be related to its antimicrobial,8–12 antioxidant13–15 and immunomodulator16 effects. Also, it might be hypothesized that honey, through increasing TNF-␣ and IL-120 might increase the levels of G-CSF. The incidence of febrile neutropenia observed in our study (> 40%) is high as compared with high income countries. The low
socioeconomic status of our patients with the prevalence of illiteracy and poverty probably contributes to this high rate. The possible positive effects of honey on hemoglobin (Hb) % and platelet count (PLT) were also evident during this study. Significant increases in Hb% and PLT were observed as a result of the intervention in group A. Conversely, significant reductions in Hb% and PLT were observed after honey withdrawal in the I/C group B. By performing a systematic search of PubMed using the terms honey, leukemia, cancer, chemotherapy, neutropenia, we only identified the study of Zidan et al.22 . In this study a type of honey called Life-Mel Honey (LMH) was found to be effective in reducing the incidence of anemia in 64% of the patients and decreasing the incidence of severe neutropenia (absolute neutrophils <500/mm3 ) by 40%. This study differed from our study in three main aspects: first, LMH is a honey that is produced by honey bees fed medicinal plants, whereas the honey used in our study was an Egyptian clover honey (T. alexandrinum L; Fam.: Leguminosae) that is produced by bees that feed primarily on the nectar of clover plants. Second, the study of Zidan et al. compared a group of patients before and after honey intake, whereas our study was a crossover study, which enabled a comparison of the effects of honey in two comparable groups of patients and in the same group of patients. Third, in the study of Zidan et al., the honey was given in a dose of 5 g/day for only 5 days, whereas in our study, honey was given in a larger dose; 2.5 g/kg twice weekly for 12 weeks for a total duration of 24 days. Despite these differences the possible positive effects of honey found in our study were nearly comparable with the results of Zidan et al. In our study, the incidence of severe neutropenia (<500/mm3 ) decreased from 47.5 to 25% as a result of the intervention in both groups combined (40 patients).
M.A. Abdulrhman et al. / Complementary Therapies in Medicine 25 (2016) 98–103
During the intervention periods, adverse effects in the form of abdominal pain, vomiting, diarrhea, fever, followed honey intake in 10 (22.7%) patients from both groups combined. These adverse effects may represent the misleading reactions to honey administration, which are benign reactions in the form of a transient aggravation of symptoms and signs of underlying illnesses.25 These reactions usually resolve spontaneously and thought to be a result of the immunomodulatory effects of honey.16,26 5. Study limitations The sample size is small. Follow up was limited to three months and the study was not blinded. A washout period was not calculated in this study because there is not yet available test to measure honey or its components in blood, and hence the half-life of honey concentration in the blood cannot be known. Therefore, as a limitation to this study, the possible carry-over effects of honey may yield statistical bias. Also a further chemical analysis of the honey used in the study to measure the phenolic content and other elements was not done. 6. Conclusion Honey intervention in a group of children with ALL may reduce the frequency of FN, the number of patients admitted with FN and the duration of their hospital stay. Further studies that include a larger number of patients are recommended to confirm that honey, has beneficial effects as a complementary agent, in children with ALL. Conflicts of interest None Contributors All authors contributed in designing the work, acquisition, analysis, and interpretation of data, drafting and revising the manuscript, and gave final approval of the version to be published. Acknowledgments All authors declare no support from any organization for the submitted work, no financial relationships with any organizations that might have an interest in the submitted work, and no other relationships or activities that could appear to have influenced the submitted work. The patients in the current study are among those followed up and admitted freely in the Children‘s hospital, Faculty of Medicine, Ain Shams University. We thank very much all patients and their parents for their participation in this study and their cooperation throughout the study periods. We also thank Ahmad M Sheded, lecturer of pediatrics, for statistical analysis. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ctim.2016.01. 009.
103
References 1. Yadav SP, Dua V, Sachdeva A. Sepsis is a major barrier to improving survival in childhood acute lymphoblastic leukemia in the developing world. J Pediatr Hematol Oncol. 2011;33(8):636. 2. Hughes WT, Armstrong D, Bodey GP. 2002 guidelines for the use of antimicrobial agents in neutropenic patients with cancer. Clin Infect Dis. 2002;34(6):730–751. 3. Sezgin G, Acipayam C, Ozkan A, Bayram I, Tanyeli A. Meropenem versus piperacillin-tazobactam as empiric therapy for febrile neutropenia in pediatric oncology patients. Asian Pac J Cancer Prev. 2014;15(11):4549–4553. 4. Ziino O, Tucci F, Rossi MR. Outpatient management of febrile neutropenia in children with cancer. Pediatr Rep. 2011;24:e8, 3 (1). 5. Castagnola E, Paola D, Giacchino R. Economic aspects of empiric antibiotic therapy for febrile neutropenia in children with cancer. Support Care Cancer. 1998;6(6):524–528. 6. Blay JY, Chauvin F, Le Cesne A. Early lymphopenia after cytotoxic chemotherapy as a risk factor for febrile neutropenia. J Clin Oncol. 1996;14:636. 7. Lockhart PB, Sonis ST. Relationship of oral complications to peripheral blood leukocyte and platelet counts in patients receiving cancer chemotherapy. Oral Surg Oral Med Oral Pathol. 1979;48:21. 8. Molan PC. The antibacterial activity of honey. 1. The nature of the antibacterial activity. Bee World. 1992;73:5–28. 9. Molan PC. The antibacterial activity of honey. 2. Variation in the potency of the antibacterial activity. Bee World. 1992;73:59–76. 10. Steinberg D, Kaine G, Gedalia I. Antibacterial effect of propolis and honey on oral bacteria. Am J Dent. 1996;9:236–239. 11. Molan PC. Honey as an antimicrobial agent. In: Mizrahi A, Lensky Y, eds. Bee Products: Properties, Applications and Apitherapy. London: Plenum; 1997:pp27. 12. Theunissen F, Grobler S, Gedalia I. The antifungal action of three South African honeys on Candida albicans. Apidologie. 2001;32:371–379. 13. Frankel S, Robinson GE, Berenbaum MR. Antioxidant capacity and correlated characteristics of 14 unifloral honeys. J Apic Res. 1998;37:27–31. 14. Gheldof N, Engeseth NJ. Antioxidant capacity of honeys from various floral sources based on the determination of oxygen radical absorbance capacity and inhibition of in vitro lipoprotein oxidation in human serum samples. J Agric Food Chem. 2002;50:3050–3055. 15. Gross HB, Polagruto JA, Zhu QY, et al. Effect of honey consumption on plasma antioxidant status in human subjects. Pap Am Chem Soc. 2004;227:U29. 16. Majtan J. Honey: an immunomodulator in wound healing. Wound Repair Regen. 2014;22(2):187–192. 17. Morales P, Haza AI. Antiproliferative and apoptotic effects of Spanish honeys. Pharmacogn Mag. 2013;9(35):231–237. 18. Abubakar MB, Abdullah WZ, Sulaiman SA, Suen AB. A review of molecular mechanisms of the anti-leukemic effects of phenolic compounds in honey. Int J Mol Sci. 2012;15:15054–15073, 13 (11). 19. Woo KJ, Jeong YJ, Park JW, Kwon TK. Chrysin-induced apoptosis is mediated through caspase activation and Akt inactivation in U937 leukemia cells. Biochem Biophys Res Commun. 2004;24:1215––1222, 325 (4). 20. Tonks AJ, Cooper RA, Jones KP, Blair S, Parton J, Tonks A. Honey stimulates inflammatory cytokine production from monocytes. Cytokine. 2003;7:242–247, 21 (5). 21. Bendalla LJ, Bradstockb KF. G-CSF: from granulopoietic stimulant to bone marrow stem cell mobilizing agent. Cytokine Growth Factor Rev. 2014;25(4):355–367. 22. Zidan J, Shetver L, Gershuny A, et al. Prevention of chemotherapy-induced ANCutropenia by special honey intake. Med Oncol. 2006;23(4):549–552. 23. Bogdanov S, Book of Honey. Honey Composition (Chapter 5). Bee Product Science. September, pp. 10. www.bee-hexagon.net; 2009 Accessed October 10.10.14. 24. Crane E. A Book of Honey. Oxford, U.K: Oxford University Press; 2016:1980. 25. Abdulrhman M. Lessons from honey. In: Paper Presented at the First International Conference on Api Phytotherapy, May 18–20. 2012. 26. Bogdanov S, Book of Honey. Honey for Nutrition and Health (Chapter 7). Bee Product Science. September 2009: pp. 10. www.bee-hexagon.net; 2014.