The role of platelet antibody and bone marrow in adult denguehemorrhagie fever with thrombocytopenia
Taufiq M. Waly*, Karmel L. Tambunan+, R.H.H. nelwan#, Hediman T.P.#, Hendarwanto#, Muljono++, Iskandar Zulkarnain#.
Abstrak
Taufiq M. Waly*, Karmel L. Tambunan+, R.H.H. nelwan#, Hediman T.P.#, Hendarwanto#, Muljono++, Iskandar Zulkarnain#.
Abstrak
Kasus demam berdarah dengue (DBD) dewasa terus meningkat pada saat ini, Dilain pihak penelitian yang dilaporkan terutama terjadi pada anak – anak. Selain itu, patofisiologi dan penatalaksanaan DBD pun masih kontroversional dan sering diperdebatkan. Telah dilakukan pemeriksaan aspirasi susmsum tulang, antibodi trombosit dan D – Dimer pada 29 pasien DBD dewasa (≥16 tahun). Diagnosis DBD ditegakkan berdasarkan kriteria klinis WHO, pemeriksaan imunokromatolografi cepat dan diperkuat dengan pemeriksaan test IH, untuk memperjelas patogenesis trombositopenia dan patofisiologi DBD dewasa. Selain itu, dilakukan pemberian steroid dengan dosis imunosupresif selama 5 hari berturut – turut pada pasien tertentu secara double blind. Hasil penelitian menunjukan bahwa pada demam / sakit < 5 hari, proses di perifer lebih berpengaruh untuk menyebabkan trombocytopenia daripada penekanan sumsum tulang, dengan antibodi trombosit sebagai faktor yang paling berpengaruh (P=0,0112). Sedangkan pada demam atau sakit lebih ≥ 5 hari tidak didapatkan variable yang berpengaruh secara statistik. Untuk steroid, bila diberikan pada demam / sakit <5 hari ternyata dapat mengurangi lama trombositopenia (<0,05), sedangkan pada demam /sakit ≥ 5 hari ternyata hanya untuk mencegah penurunan tajam dari jumlah trombosit (P<0,01). Kesimpulan: trombositopenia pada orang dewasa terutama disebabkan oleh karena proses perifer ketimbang penekanan megakariosit sumsum tulang. Pada demam/sakit <5 hari, steroid memperpendek lama trombositopenia dan mencegah penurunan tajam dari jumlah trombosit. Akan tetapi pada demam/sakit ≥5 hari steroid hanya mencegah penurunan tajam dari jumlah trombosit.
Abstract
Adult dengue hemorrhagic fever (DHF) cases are increasing rapidly, while most of the studies reported chidren cases. Controversies in DHF pathophysiology and treatment are still a matter of debate. In this study, bone marrow puncture, platelet antibody and D-Dimer examination were caried out in 29 adult DHF patientes (more than 16 years). The diagnosis of DHF was confirmed clinically using the WHO criteria, rapid immunochromatolographic examination supported by hemagglutination inhibilition (HI) test to clarify the pathogenesis of trombocytopenia and the pathophysiology of DHF in adult patient. Besides that, steroid was given in immunosuppressive dose during 5 days to selected patients (on double blind basis). The cross sectional study, cohort study and double blind trial were the methods used in this study. The results showed that peripheral processes apparently were more affected than bone narrow supression in causing thrombocytopenia, in less than 5 days fever group, with platelet antibody as a leading factor (P= 0.0112). While, in the more than 5-day fever group, there are no significant variables. If steroid was given, in the less than 5-day fever group, apparently the length of trombocythopenia could be decreased (P <0,05), while thrombocyte level could be kept from lowering too fast, in the more than 5 day fever group(P<0,1). Conclusion: Most of thrombocytopenia was caused by periferal process than bone marrow suppression. In the less than five day fever group, the steroid could decrease the length of thrombocytopenia and keep the thrombocyte level from lowering too fast. However, in the more than 5 day group it only kept the thrombocyte level from lowering too fast.
Keywords : Reticulo endothelial system, steroid, immunologic reaction
* Departement of internal medicine, Faculty of Medicine University Of Indonesia / Cipto
Mangunkusumo Hospital
+ Division of Hematology- Medical OncologyDepartement of internal Medicine, Faculty of Medicine, University of Indonesia / Cipto mangunkusumo Hospital
# Division of tropical and Infectious Diseases, Departement of internal Medicine, Faculty of Medicine Universityof Indonesia /Cipto MangunkusumoHospital.
++ Departement of Pediatrics, Faculty of Medicine University of Indonesia/ Cipto Mangunkusumo Hospital.
Dengue Hemorrhagic Fever (DHF) is still an endemic illness in Indonesia. Today, DHF is endemics 26 provinces (except Bengkulu).1 In one of the surveys done by the Health Departement of the Republic of Indonesia at nine cities, it was showed that Aedes Aegypty was present in one third af the houses or public places investigated.2 The lates data from the health Departement of the Republic af Indonesia, Aedes Aegypty was found in more than 10 % of the houses and buildings in 27 provinces.1 Althougt national DHFmortality rate is suppressed until 2.5%. mortality rate in hospitals caused by DHF mortality rate is still high (5.7%).1 In the other hand, the mordibity rate of adult DHF patients apparently was increased from 4.3% in 1968 to 40.56% in1996.1
Thrombocytopenia is a central issue in the oathogenesis of DHF and bone marrow plays a role to cause that. It is supposed thet suppression of bone marrow, especially the megakariocytes can cause thrombocytopenia in DHF patients with less than 5-day fever. While in patients with more than 5-day fever thrombocytopenia is caused by peripheral process.3.4 The spread of virus or viral antibody complex to bone marrow, was proved by rothwell et al.5 Bone marrow assessment data in adult DHF patient (more than 16 years) is still lacking. Most of bone marrow studies were carried out in children (less than 16 years).6.7.8.9.10.Nelson and Bierman carried out bone marrow axaminations in adult patients with hemorrhagic faver (not just dengue), but they did not explain whether the results were similar to DHF or not.11 Clinical symptoms in children DHF patients are more dangerous than in adults, thrombocytopenia was found only in 53,9% and shock in 14 % of patients.12 However, in adult, thrombocytopenia was found in 82 % and shock in 14 % of patients.13 in 1995, the mortality rate for DHF at Cipto Mangunkusumo Hospital showed the same feature, i.e. 3.6 % in children and 0,8 % in adult. Furthermore, there is no statistical analysis to determine the decrease of megakariocytes in bone marrow as a main cause of trombocytopenia. Other causes of thrombocyte destruction in the periphery are disseminated intravascular coagulation (DIC), reticuloendotherial system (RES) activity, viral activity and immunologic reaction to destory the immune complex that was adhered to the thormbocyte. Buthep and friends explained thet endothelium destruction was a main cause of thrombocytopenia and leucopenia.14 Funahara said that there were three leading peripheral factors to cause thormbocytopenia I.e. virus that destroyed the thrombocyte directly, endothelium destruction / comsumtive coagulopathy and antigen antibody complex that destroyed the thrombocyte. 15 Radiosotope technology proved that RES activity was also a leading factor to cause thrombocytopenia.16
Platelet antibody can be positive in a DHF patien and played a rol in causing thrombocytopenia. This is due to the increase of stimulation in autoreactive lymphocytes which is always present in human causing autoimmune reaction to thrombocyte. IN the normal situation. Body immununoregulator system always succeeds to prevent autoreactive lymphocythes in causing autoimmune reaction (normal immunology homeostasis mechanism) lsarangkura and Tuchinda proved that thrombocyte transfusion to DHF patients with severe thrombocytopenia showed no significan result.17
Until now. it is supposed that the thrombocytes are not the target cell of dengue virus.The target cells of dengue virus are monocytes macrophages and Kupffer cells. 18 Therefore, the adhesion of immune complex (virus antibody complex) to endothelial wall, thrombocytes, and kidnay cells cells, event o brain and pancreas cells was more likely to be body over reaction.19.20 Halsted, sthe leader of the secondary heterologous infection theory, changed his option to genetic theory to explain the severe clinical sympptom that could happen in primary infection, occurred in Eourope, United States and Australia. 21,22 Myomin and Futrakul gave steroid to dangue shock syndrome (DSS) patients and got a good result, but Sumarmo in his two studies (1973 and 1978) did not find a significant result.23,24
In this study, we tired to give steoroid to the patients. This is according to aa assumption that in an adult DHF patient the bone marrow was less affected to the peripheral process (immunologic reaction). Besides that, the possibility of positive platelet antibody and hypersensitivity reaction stage III as DHF basic pathophisiology still needs a consideration.
METHODS
Sudy design
1. Cross sectional study. This design is useful to find out the role of bone marrow and the influence of the other variables in causing low thrombocyte count at the time of admission.
2. Cohort study. This is useful to evaluate thrombocytopenia until it reaches the normal level and the influence of platelet antibody and steroid on thrombocyte count.
3. Double blind test. This is useful to know the effect of steroid on thrombocyte count.
Inclusion criteria to his study
Patients over 16 years which were clinically proved as DHF, according to the WHO criteria, except the increase in hematrocrit value. In the WHO criteria,²⁵ the increase in hematocrit value is based on the increase above the normal value according to age, sex and population. In Indonesia, there is no data about that.
The Inclusion criteria are: fever (or history of acute fever) less than 8 days (2 – 7 days), thrombocyte level <100.000/mm3, hemorrhagic tendencies (minimally rumple Leede) and other criteria to prevent bias thrombocytopenia i.e. normal spleen (physical finding), chest x-ray, and urine analysis
The patients were divided into 2 groups, the group with fever less than 5 days and the group with fever more than 5 days.
Exclusion criteria for this study
Patients over 16 years which were clinically proved as DHF, according to the WHO criteria, except the increase in hematrocrit value. In the WHO criteria,²⁵ the increase in hematocrit value is based on the increase above the normal value according to age, sex and population. In Indonesia, there is no data about that.
The Inclusion criteria are: fever (or history of acute fever) less than 8 days (2 – 7 days), thrombocyte level <100.000/mm3, hemorrhagic tendencies (minimally rumple Leede) and other criteria to prevent bias thrombocytopenia i.e. normal spleen (physical finding), chest x-ray, and urine analysis
The patients were divided into 2 groups, the group with fever less than 5 days and the group with fever more than 5 days.
Exclusion criteria for this study
All patients with negative IgG or IgM on rapid immunocchromatography, or patients whose second hemagglutination inhibition (HI) test level did not increase four times compared to the first HI test or whose HI test was not more than 1/2560 (presumptive positive) are excluded from this study.
Double blind test grouping
We divided all samples into two groups, i.e. the group of patients receiving steroid and fluid treatment(10 patients) and patients receiving steroid and placebo. Then we analyzed the results in the group receiving steroid compared to the group receiving placebo, in the less than 5-day and more than 5-day fever groups.
Exclusion criteria for steroid administration
The patients with gastrointestinal bleeding and whose thrombocyte level is lowered to 20.000 / mm3 in a routine examination was excluded from receiving steroid.
Investigation procedure
The Patients were subjected to bone marrow puncture and blood collection for haematologic, serologic and D-Dimer investigation and viral isolation. Some patients were subjected to steroid administration scheme.
The increase of trombocyte level in all patients was observed day by day. The patient was sent home if the thrombocyte level was more than 100.000/mm3 in two subsequent days; but before that, the blood sample should be taken for the second HI test. For a patient with positive D-Dimer, the blood sample would be taken everyday until we found negative D-Dimer.
Laboratory procedures
Rapid immunochromatography
The principal is to know if there is any antibody to dengue virus types 1,2,3,4 in the patient. We used the patients serum (from 3ml of venous blood) and applied it on nitrocellulose paper containing dengue virus antigen. Positive rapid immunochromatography test showed the change of color of control IgG and/ or IgM box. The instrument was adjusted to the WHO criteria based on positive III test for IgG and IgM. Therefore, a secondary infection showed changes of color in the IgG and IgM boxes.
Assessment of bone marrow cell
The bone marrow puncture (BMP) was carried out as fast as possible, in 24 hours or more, and before the administration of steroid. The bone marrow megakariocytes were calculated and its morphology were analyzed. The number of megakariocytes is considered normal when we found 12 – 50 megakariocytes/50 low magnification field. This is the normal value used in the Division of Hematology, Departement of internal Medicine, CiptoMangunkusumo Hospital. According to other investigators the bottom limits are 0 or 11 and the upper limits are 27 or 38.²⁶⁻³⁰ Besides that, the differential counts of 200 bone marrow cells were done.
Hematologic investigation
We took 2 1/2ml of venous blood and added 2mg of EDTA. The sample was sent for complete blood count I.e. Hb, Ht, and leucocyte and thrombocyte count, immediately to the clinical pathology Laboratory, Cipto mangunkusumo / Persahabatan Hospital (for patients who came at night) or to the hematology Laboratory, Cipto Mangunkusumo Hospital for patients who came in the morning. The instruments used were Coulter Max M (Clinical Pathology Laboratory) and Micro Coba S (Hematology Laboratory).
Serologic Investigation
1. Inhibition hemagglutination test. This test Was based on the fact that dengue virus had the ability to agglutinate red blood cells, and the specific antibody in the patients serum would inhibit the agglutination process. We took 3ml of venous blood, put it in vucatainer tube and sent the samples immediately (within 4 hours) to the Health Research and development Institution of NAMRU – 2 Laboratory. The HI test were conducted when the patients came and leaved the hospital. The positive result was shown by a four – time increase in HI level.
2. Platelet antibody examination was carried out by immunofluorescent method.
D-Dimer Investigation
We took 4,5 ml of venous blood, added 1 ml of 3.8% sodium citrate and the samples were centrifuged (using Rcsorval 3 type centrifuge) at 4’C and 2000 x g for 10 minutes. The plasm was put into chromotimer Behring. This investigation was carried out at the Hematology laboratory Cipto Mangunkusumo Hospital.
Viral isolation
Viral isolation of dengue virus was carried out at NAMRU – 2 or the Microbiology Laboratory. University of Indonesia. The Procedure was: patients serum inoculation to the intrathorax of female/ male Aedes aegypty, incubation between 10 – 13 days at 30’ C with relative humidity, and detection of dengue antigen. The detection of dengue antigen was done at the end of incubation time, I.e. the mosquito was killed and the head of the mosquito was reacted againts anti – dengue antibody which was conjugated to a fluorescence dye (isotiocyanat) and the screened using fluorescence microscopy. The presence of the virus grow could be seen as green-yellow flourecent particle.
To isolate the dengue virus, we took 3 ml of venous blood and put in into a vacutainer tube and sent it to the lab in less than 4 hours. Viral isolation was stopped if most of the results from 30 patients were negative.
Steroid administration scheme
The steroid used was Medixon® (metil prednisolon) in immunosuppressive dose, 1.2 – 1.5 mg / kg body weight/day, during 5 days.
Statistical analysis methods
Statistical analysis (chi-square, thrombocytopenia relative risk and multiple regression analysis) was performed using SPSS program for Windows (6.0)
RESULTS
There were 33 cases from May through July 1997, whish were collected in Cipto Mangunkusumo and persahabatan General Hospital, and 29 of them were included in the study. The entire result of this study is shown in Tables 1,2,3,and Figure 1
Table 1. The result of bone marrow assessment
Table 2. Result of hematologic investigation
Table 3. The result of physical finding, serologic and D-Dimer investigations, viral isolation, and steroid administration
DIC = Disseminated intravascular coagulation
DIC NOT Com = decompensated DIC
Figure 1. The length of throbocytopenia in the 2-4 day fever and 5-7 – day fever group
LT = the length of thrombocytopenia, AbTr = platelet antibody, str = steroid
Figure information :
LT of 27 patients
1. Average
2. AbTr (+), Str (+)
3. AbTr (-), Str (-)
4. Abtr (-), Str (+)
5. AbTr (+), Str (-)
2-4 day fever group
6. Average
7. AbTr (+), Str (+)
8. Abtr (-), Str (-)
9. Abtr (-), Str (+)
10. Abtr (+), Str (-)
5-7 day fever group
11. Average
12. AbTr (+), Str (+)
13. Abtr (-), Str (-)
14. Abtr (-), Str (+)
15. Abtr (+), Str (-)
The thrombocyte count decreased exessively in four patients with positive platelet antibody, after they achieved 100.000/mm3. Those four patients did not get steroid. They represented 40% 0f the 10 patients with positive platelet antibody and without steroid administration.
There were four patients with trombocytopenia that lacked variable causing thrombocyteopenia (negative platelet antibody, normal megakariocyte in bone marrow assessment, no fever or viremia, no bleeding, and negative DIC).
Statistical Analysis
Multiple Regression Analysis
Multiple regression analysis was used to analyze the variables influencing the first thrombocyte count (TR 0). In the less than 5-day fever group, all variables (platelet antibody, decrease in bone marrow megakariocytes, decompensated DIC, and real bleeding) showed significant influence (P <0,05) on TR 0, with platelet antibody as a leading factor (P=0.0112). The order of the influencing variable platelet antibody, decompensated DIC, real bleeding and decrease in bone marrow megakariocytes. However, multiple regression analysis to analyze the variable influencing TR 0 in the more than 5 day fever group show that no variable had significant influence. Furthermore 1 way multiple regression analysis on the entire patient (29 cases) show that apparently only platelet antibody had significant influence on TR 0 (P=0,09)
Trombocytopenia relative risk analysis.
The relative risk of thrombocytopenia in patient with platelet antibody (+) was increased 1,5 , 1,06 , and 1,75 times (confidence limits 95%) in the less than 5 days fever group, more than 5 day fever group and in entire patient respectively compared to the patient with negative platelet antibody.
Chi-square analysis
Chi-square analysis show a significant difference in the duration between patient with and without steroid administration (P<0,05), in the less than 5 day fever group. While in the more than 5 day fever group there was no significant difference between patient with and without steroid administration. One way chi-square analysis show that platelet antibody might be the cause of the fast decrease in thrombocyte level (P=0,0568). Furthermore, one way chi-square analysis show that steroid administration to patient with positive platelet antibody in the two groups might keep the thrombocyte level from lowering too fast (P= 0.07364).
DISCUSSION
In the less than 5-day fever group, positive platelet antibody was found in 62.5 % of the cases and the quite high influence of decompensated DIC to cause early thrombocytopenia, lead to the conclusion that positive platelet antibody and decompensated DIC could be regarded as a leading factor in causing thromboytopenia. Furthermore, by seeing the strength of positive platelet antibody influence to cause early thrombocytopenia in the less than 5-day fever group (P= 0.00012) and the higher relative risk in developing thrombocytopenia in this group (1,5 times). We concluded that the bone marrow suppression was not the main cause of thrombocytopenia.
The role of steroid administration in the less than 5-day fever group to decrease the duration of trombocytopenia was evaluated, even it was only given in two patients. Actually, there were four patients who received the steroid in this group, but two patients were excluded from the study, because the result of rapid immunochromatographic and HI tests were negative, even, obviously the clinical symptoms were DHF. The negative serology in those two patients needed further consideration beause steroid administration was given too early (in the first and second day of the illness). This condition could inhibit the formation of IgG and IgM, resulting in negative serology. Steroid administration decrease the length of thrombocytopenia significantly (P=0,01850.
In the more than 5-day fever group, there were no variables that influence the first thrombocyte count (the number of thrombocytes at the time before BMP). However, RES activity needed to be considered as the cause early thrombocytopenia in this group. This was based on the four patients who developed thrombocytopenia without influencing variables causing thrombocytopenia. The average thrombocyte count in those four patients was 33.000. The occurrence of increased RES activity to destroy the thrombocytes that were invested by immune complex was already proved by radioisotope. On the other hand, the low relative risk of the patients with positive platelet antibody compared to negative platelet antibody to develop thrombocytopenia, was due to the immunoregulator that secceeded to neutralize the autoreactive lymphocytes, after 5 days. The decreased of autoreactive lymphocytes was caused by the decrease of the stimulation caused by the immune complex, because a lot of immune complex were destroyed. Furthermore, the condition of positive platelet antibody became negative, as a consequence of the decrease in autoimmune reaction.
In the more than 5-day fever group, steroid had no effect on the duration of thrombocytopenia. Steroid administration only slowed down the lowering of trombocyte level. This was due to the inhibition effect of steroid in the immunoregulator system. Steroid inhibited autoreactive lymphocytes, i.e, T helper cell (Th) to give wrong impulse to B lymphocytes. Time was needed to take over the regulation of Th lymphocyte (autoreactive lymphocyte) by the steroid, so that duration of trombocytopenia was not decreased. In the other hand, steroid was statistically proved to keep the thrombocyte level from lowering too fast which was caused by presence of platelet antibody.
CONCLUSION
1. Peripheral process was more affected that bone marrow suppression to cause thrombocytopenia in the less than 5-day fever group, with platelet antibody as a leading factor.
2. In the more than 5-day fever group, the influence of RES activity as a cause of thrombocytopenia needs to be considered.
3. Steroid administration in the less than 5-day fever group could shorten the duration of thrombocytopenia, but in the more than 5-day fever group, the effect was only to keep thrombocyte level from lowering too fast.
REFERENCES