Changes in splenic volume after partial splenic embolization in children

Changes

in Splenic Volume After Partial Embolization in Children By Yasuhiro

Watanabe, Takuji Todani, Kagawa, Japan

l Aim: The authors studied changes in splenic volume after partial splenic embolization (PSE) in children who had thrombocytopenia caused by splenomegaly. Materials and Methods: Five children with hypersplenism (ages of 1, 6, 7, 10, and 14 years) underwent PSE. The follow-up period ranged from 1 month to 8 years 9 months. Whole and infarcted splenic volumes were measured from computed tomography images obtained before and after PSE. Resu/ts: Within 2 weeks after PSE, the whole splenic volume increased to 110% to 140% of the pre-PSE volume. The infarcted area disappeared 3 to 4 months after PSE, and the whole splenic volume decreased markedly. The decreased volume tended to be stable during the follow-up period. The splenic enlargement before PSE ranged from 7.2 to 14.2 times the standard splenicvolumefor a corresponding body weight. The splenic volume did not become normal after PSE; it remained between 2 and 7 times larger than standard. Platelet counts stayed above 10 x 104/mm3 when the splenic volume reduced to 2 times the standard volume. Conchrsion; The first 4 weeks after PSE is the most dangerous period for the development of splenic rupture because of marked expansion of the splenic capsule. The infarcted parenchyma is absorbed 3 to 4 months after PSE, and the whole splenic volume is reduced. Thereafter, the splenic volume is stable for several years. Reembolization is not likely to be necessary in the majority of patients who have had splenic embolization with an infarction rate of 280%. Copyright @ 1996 by W.B. Saunders Company INDEX WORDS: hypersplenism.

Partial

splenic

embolization,

splenic

volume,

P

ARTIAL SPLENIC EMBOLIZATION (PSE) is a safe and effective treatment for control of hypersplenism in children with portal vein thrombosis and in long-term survivors who have undergone the Kasai operation for biliary atresia. Fulminant septicemia, a lethal complication after splenectomy, which occurs more commonly in younger children, could be avoided by PSE. Although PSE requires longer hospitalization than does splenectomy,r the former preserves some amount of functioning splenic parenchyma. PSE is an attractive alternative to splenectomy in childhood. Little is known regarding changes in splenic volume after PSE.z We measured splenic volume before and after PSE by analyzing computed tomography (CT) images. MATERIALS

AND METHODS

Five children with hypersplenism (4 with bihary atresia after the Kasai operation, 1 with thrombosis of the prehepatic portal vein) underwent PSE. Their ages at the time of the procedure were 1,6, Journa/offediatricSurge~,

Vol31,

No 2 (February),

1996: pp 241-244

and Takuo

Splenic

Noda

7, 10, and 14 years. The follow-up period (with evaluation by CT) ranged from 1 month to 8 years 9 months. Whole and infarcted splenic volumes were measured from enhanced CT images obtained before and after PSE. The outline of the spleen and infarcted foci on each CT image was delineated, and the marked area was measured with a computer image analyzer, IBAS-2 (Kontron, Munich- Germany). Using scale marks on the image, the area values were converted into the actual area (in square centimeters). The splenicvolume (in cubic centimeters) was calculated by summing up the area and then multiplying by slice thickness (in centimeters). The infarction rate was calculated by dividing the infarcted volume by the whole splenic volume; this was expressed as a percentage. The ratio of splenic volume to body weight (cm/kg) was used to exclude physical growth. The volume of splenic enlargement compared with a standard splenic volume for a corresponding body weight also was determined. The standard volume was calculated using the formula y = 6.11 r”sl (y, standard splenic volume in cubic centimeters; X, body weight in kilograms), which was derived by us from 44 normal Japanese children and young adults (age range, 9 days to 25 years) (Fig 1). Each splenic volume was calculated from CT images as described above. RESULTS Patient

1

A 12-year-old boy (Fig 2) underwent the Kasai operation at SOdays of age. Portal hypertension and hypersplenism developed (white blood cell [WBC] count, 3,0001mm3; red blood cell [RBC] count, 470 x 104/mm3; platelet count, 4.0 x 104/mm3). The splenic volume was 1,316 cm3 (38.5 cm3/kg), which is 12.3 times larger than standard (Fig 3). PSE was performed, and the splenic volume increased by 112.7%, to 1,483 cm3 (43.4 cm3/kg), 10 days after PSE. The amount of infarcted spleen was 31.4%, and the platelet count increased to 13.6 x 104. Seventyfive days after PSE, the infarction rate decreased to 9.8%. The whole splenic volume reduced by 71.5% of the pre-PSE volume, to 940 cm3 (27.5 cm3/kg). Eight months after PSE, the infarcted area was not observed on enhanced CT images. Although the

From the Depanment of Pediatric Surgery, Kagawa Medical School, Kagawa, Japan. Presented at the 28th Armuai Meeting of the Paci$c Association of Pediattic Surgeons, Huatulco, Oaxaca, Mexico, May 14-18, 1995. Address reprint requests to Yasuhiro Watanabe, MD, Depanment of Pediattic Surgety, Kagawa Medical School, 1750-13 Miki, Kitagun, Kagawa 761-07 Japan. Copyright 0 1996 by W B. Saunders Company 0022-3468l96l3102-0007$03.00l0 241

242

WATANABE,

~=6.11

Splenic Volume (Crrf ‘1 400

x0.8'

r=o.9274

Chcmg;in

r-r=44

TODANI,

AND

NODA

splenic volume

x

350

x

300 *so 200 150 too so 0 0

IO

*o

io

4b

Body Weight

70

io

io

(kg)

Fig I. Standard splenic volume as a function standard splenic volume in cubic centimeters; kilograms). The formula was derived from 44 children and young adults, whose age range was The splenic volume was calculated from CT images Materials and Methods section.

of body weight ( y, x, body weight in normal Japanese 9 days to 25 years. as described in the

whole splenic volume was reduced further, to 905 cm3 (22.7 cm3/kg), the platelet count decreased to 7.8 x 104. Therefore, PSE was performed again. The infarction rate estimated by angiographic findings during the procedure was 60%. However, 10 days after the second PSE, CT showed that only 10% of the spleen was infarcted. The whole splenic volume decreased gradually to 16.8 cm3/kg (43.7% of the pre-first PSE volume per body weight) by 21 months after the first PSE (13 months after the second PSE). Three years and 1 month after the first PSE, the whole splenic volume was 15.2 cm3/kg (39.4% of pre-first PSE volume per body weight). This value has been stable for up to 8Yd years after the first PSE (14.7 cm3/kg, 38.2% of pre-first PSE volume, and 5.1 times larger than standard). The platelet count fluctuated between 8 and 10 x 104/mm3 during this period. Platelet co”“t

Fig 3. Changes in splenic volume after PSE. A change was calculated as spienic volume + standard splenic volume. The standard splenic volume was calculated with the formula described in Fig 1.

Patient 2 A baby girl who had undergone the Kasai operation age 64 days (Fig 4) showed moderate hypersplenism at 17 months of age (WBC count, 5,900; RBC count, 359 x 104; pIatelet count, 9.1 x 104), The splenic volume was 306 cm3 (29.5 cm3/kg), which is 7.5 times larger than the standard volume (Fig 3). PSE was performed with an 81.9% infarction rate, and the spleen enlarged to 110.0% of the pre-PSE volume by 9 days post-PSE. The platelet count increased to 33.4 x 104. The infarcted foci were recognized on CT images obtained 40 days after PSE, and disappeared 4 months after PSE. The splenic volume was 10.3 cm3/kg (35.1% of the pre-PSE volume per body weight). The splenicvolume decreased to 7.0 cm3/kg (23.7% of pre-PSE volume per body weight) 6 years 10 months after PSE (2.1 times larger than standard), and the platelet count was 15.2 x 104. Patient 3 A 14-year-old girl who had undergone the Kasai operation at age 50 days underwent PSE with a 74.4% infarction rate. The splenic volume before PSE was Plcttelet count

volume (% of ihe pre-PSE volume)

0 Fig 2. Patient 1. Changes PSE. Volume is expressed volume. PSE, partial splenic

in splenicvolume as a percentage embolization.

and platelet count after of the pre-PSE splenic

,

2

s

4

Follow-lip

Fig 4. PSE.

Patient

2. Changes

in splenic

s

6

(years)

volume

and platelet

count

after

SPLENIC

VOLUME

AFTER

243

PSE

1,071 cm3 (20.8 cm3/kg), which is 7.2 times larger than standard. The spleen enlarged to 1,477 cm3 (137.9% of the pre-PSE volume) 11 days after PSE. The platelet count increased to 20 x 104. Forty-three days after PSE, the splenic volume decreased to 633 cm3 (14.6 cm3/kg), which is 70.1% of the pre-PSE volume. The infarcted foci were still apparent on CT images. Patient 4 Hypersplenism developed in a girl aged 6 years 3 months (Fig 5) who had thrombosis of the prehepatic portal vein (WBC count 2,200; RBC count; 448 x 104; platelet count, 4.8 x 104). The splenic volume was 485 cm3 (30.3 cm3/kg), which is 8.4 times greater than the standard volume before PSE (Fig 3). PSE was performed with an infarction rate of 62.5%, and the volume increased to 660 cm3 (41.2 cm3/kg) 16 days after PSE. The splenic enlargement was 136.1% of the volume before PSE. Seventeen months after PSE, the splenic volume decreased to 13.6 cm3/kg (45.0% of the pre-PSE volume per body weight). However, 21 months after PSE, the spleen enlarged again, to 18.4 cm3/kg (60.7% of the pre-PSE volume per body weight), which is 5.3 times larger than standard. The volume changes appeared to become stable 2 years 5 months after PSE (17.0 cm3/kg, which is 56.1% of the pre-PSE volume per body weight, and 5.0 times larger than the standard volume). The platelet count decreased to 6 x 104. Patient 5 A girl aged 10 years 7 months who had undergone the Kasai operation at age 65 days showed hypersplenism (WBC count, 1,500; RBC count, 478 x 104; platelet count, 2.2 x 104). The splenic volume was 1,310 cm3 (45.8 cm3/kg)-14.2 times larger than standard. PSE was performed with a 75.4% infarction rate, and the spleen enlarged by 124.8%, to 1,636 cm3 (57.2 cm3/kg), 8 days after PSE. Six months after PSE, the splenic volume decreased to 17.2 cm3/kg. This was PLatelet count volume (% of the pre-PSE volume)

ight

i Follow-up (years)

Fig 5. PSE.

Patient

4. Changes

in splenic

volume

and platelet

count

after

37.6% of the pre-PSE volume per body weight, and 5.4 times larger than standard. The platelet count was 8.3 x 104. DISCUSSION

Partial splenic embolization is now widely accepted as a safe and effective alternative to surgical splenectomy,3,4 and is frequently used in children who have hypersplenism or splenomegaly.1~5~6 Hematological improvement can be maintained for a long time when an adequate amount of spleen is embohzed.4x5 Accurate assessment of the embolized splenic volume is essential in patients who have undergone PSE. Ultrasonography,7 single-photon emission CT,8 and axial CT9-i1 have been used for volume assessment. CT appears to be simple and accurate.12 Splenic volume in patients with hypersplenism is variable, and also depends on age, especially in childhood. Standardization of volume assessment of the spleen is necessary in the pediatric group. In the present study, the enlargement was assessed against standard volume (Figs 1 and 3). Before PSE, splenic volume ranged from 7.2 to 14.2 times larger than standard, and platelet counts ranged from 2.2 to 9.1 X 104/mm3. In our series, splenic volume of more than 7 times normal was a good indication for PSE. The spleen enlarged to 110% to 138% of the pre-PSE volume 1 to 2 weeks after PSE, probably owing to necrosis and to edema of the surrounding splenic parenchyma. The marked expansion of the splenic capsule could result in rupture. Accordingly, patients should be strictly protected against blunt trauma to the upper abdomen for approximately 4 weeks after PSE. In the present study, a decrement phase of splenic size followed the enlargement phase, owing to resorption of the necrotic tissue. The nonenhanced area indicating infarcted foci disappeared from the CT images 2.5 to 4 months after PSE. However, the splenic size continued to decrease, even after the disappearance of infarcted foci, probably because of scar formation in the spleen, and tended to stabilize 4 to 21 months after PSE. In patients with hypersplenism, a platelet count of more than 10 x 104/mm3 is considered clinically satisfactory. In our study, a greater platelet count could be achieved for a long time with a PSE infarction rate of 80%. However, the spleen size remained 2 times standard and did not decrease to normal, even after infarction of more than 80% (Fig 3). In this series, an adequate platelet count was obtained in patient 2 (with a spleen 2 times larger

244

WATANABE,

than standard). In patients 1 and 4, the spleen size was 4 to 5 times larger than standard, and the platelet count remained below 10 x 104/mm3 several years after PSE, with no clinical signs of thrombocytopenia. Splenic regeneration is observed in children after PSE.Q We also observed that splenic volume increases gradually with age. However, this phenomenon does not mean a pathological reenlargement of

TODANI,

AND

NODA

the spleen. The splenic volume after PSE tended to be stable when the volume was standardized by body weight. Although repeat embolization for the patient with subnormal platelet count is recommended,3 the second PSE usually is not necessary for at least 10 years if the first PSE has achieved sufficient splenic embolization, ie, 2 80% infarction.

REFERENCES 1. Kumpe DA, Rumack CM, Pretorius DH, et al: Partial splenic embolization in children with hypersplenism. Radiology 155:357362,1985 2. Ertzner TW, Sandier MP, Kulkarni MV, et al: Quantitative evaluation of postembolized splenic tissue using technetium-99m sulfur colloid. Clin Nucl Med 12:281-286, 1987 3. Alwmark A, Bengmark S, Gullstrand P, et al: Evaluation of splenic embolization in patients with portal hypertension and hypersplenism. Ann Surg 196:518-524,198Z 4. Shah R, Mahour GH, Ford EG, et al: Partial splenic embolization. An effective alternative to splenectomy for hypersplenism. Am Surg 56:774-777,199O 5. Brandt CT, Rothbarth LJ, Kumpe D, et al: Splenic embolization in children: Long-term efficacy. J Pediatr Surg 24:642-645, 1989 6. Israel DM, Hassall E, Culham JA, et al: Partial splenic embolization in children with hypersplenism. J Pediatr 124:95-100, 1994

7. Dittrich M, Milde S, Dinkel E, et al: Sonographic biometry of liver and spleen size in childhood. Pediatr Radio1 13:206-211, 1983 8. Kavanagh GJ, Kavanagh JT, Kavanagh PB, et al: Automated volume determination of the liver and spleen from Tc-99m colloid SPECT imaging. Quantification of the liver functional and nonfunctional tissue in disease. Clin Nucl Med 15:495-500, 1990 9. Heymsfield SB, Fulenwider T, Nordlinger B, et al: Accurate measurement

of liver,

kidney,

and

spleen

volume

and

mass

by

computerized axial tomography. Ann Intern Med 9O:lS5-187,1979 10. Breiman RS, Beck JW, Korobkin M, et al: Volume determnations using computed tomography. AJR 138:329-333,1982 11. Prassopoulos P, Cavouras D: CT assessment of normal splenic size in children. Acta Radio1 35:152-154, 1994 12. Moss AA, Friedman MA, Brito AC: Determination of liver, kidney, and spleen volumes by computed tomography: An experimental study in dogs. J Comput Assist Tomogr 5:12-14, 1981