To the editor:

Red cells of patients with hereditary spherocytosis (HS) have a decreased surface-to-volume ratio,1  leading to their trapping and destruction during their passage through the splenic cords.2-4  Therefore, surgical total splenectomy (TS) by removing the main site of red cell destruction lengthens the red blood cell life span and, according to published guidelines, should be recommended in severe cases and needs to be considered in intermediate forms of HS.5-8  However, spleen removal is associated with loss of important physiological functions: splenectomized patients are exposed to long-term complications including infectious risks (particularly high in children under 6 years of age9 ) and vascular risks.10,11  To preserve the spleen function, subtotal splenectomy (STS) has been proposed as an alternative to TS. We and others published reports regarding a series of HS patients who underwent STS and showed that it reduced the hemolytic rate and increased the red cell life span while maintaining efficient splenic phagocytic function.12-16  However, long-term results are still lacking.

We report here an update of our STS series which includes 79 patients from 71 families who underwent STS for HS at Bicêtre Hospital since February 1985 with at least 3 years’ follow-up. This represents the largest STS series in HS, with the longest follow-up to date (mean follow-up, 11 ± 0.7 years standard error of the mean [SEM]; range, 3-23 years). Patients’ characteristics are summarized in Table 1. Thirty-nine patients (group A) underwent STS for transfusion dependency (n = 31) and/or anemia with a baseline hemoglobin (Hb) level under 95 g/L (n = 8). For the 40 other patients (group B), all with a baseline Hb above 95 g/L, STS was performed for chronic fatigue (n = 18), mild asthenia with gallstones requiring cholecystectomy (n = 16) or with frank icterus (n = 6). The mean age at surgery was significantly lower in group A than in group B (4.3 vs 11 years, P < .0001). The procedure consisted of open surgical removal of 85% to 95% of the spleen, as previously described.12,13  A cholecystectomy was performed when preoperative ultrasonography demonstrated stones in the gallbladder. The postsurgical complications were: 2 early hemorrhages, 1 late bowel occlusion, 2 patients with recurrent abdominal pain requiring completion, and 5 symptomatic wandering remnants requiring TS for 1 and fixation for the others. The phagocytic function of the spleen remnant was assessed using pitted red cell count, Howell-Jolly body count, and splenic scintigraphy performed at 1-year postsurgery and then every 5 years.17  A functional remnant spleen was documented following STS in 96% of the patients. Three patients were considered as asplenic: 1 showing a very small fixation (remnant length, 1 cm), the 2 others because of an early postsurgery necrosis. No severe infections (requiring hospitalization) or thrombotic event was noted during the follow-up period.

Table 1

Main characteristics of patients from group A and B

All patientsGroup AGroup BP
No. 79 39 40  
Sex ratio, M/F 1.51 1.44 1.67 NS 
Mean age at diagnosis, mo 19 ± 4.9 8.5 ± 3.7 30 ± 8.8 
Protein Ankyrin = 18 Ankyrin = 11 Ankyrin = 7 NS 
Band 3 = 1 Band 3 = 0 Band 3 = 1 
Protein 4.2 = 1 Protein 4.2 = 0 Protein 4.2 = 1 
α-Spectrin = 2 α-Spectrin = 2 α-Spectrin = 0 
β-Spectrin = 12 β-Spectrin = 7 β-Spectrin = 5 
Mean hemoglobin, g/L 97 ± 2.2 83 ± 1.9 110 ± 2.6 **** 
Mean reticulocytes, G/L 392 ± 18 391 ± 26 393 ± 24 NS 
Mean total bilirubin, µmol/L 59 ± 6 47 ± 6.9 68 ± 9 NS 
Mean no. of transfusion per year 1.6 ± 0.4 3.1 ± 0.7 0.13 ± 0.02 **** 
Lithiasis, % 32.1 20.0 47.5 ** 
Mean age at surgery, y 7.4 ± 0.6 4.3 ± 0.6 11 ± 0.7 **** 
Mean time of follow-up, y 11 ± 0.7 12 ± 0.9 9.3 ± 1.0 NS 
All patientsGroup AGroup BP
No. 79 39 40  
Sex ratio, M/F 1.51 1.44 1.67 NS 
Mean age at diagnosis, mo 19 ± 4.9 8.5 ± 3.7 30 ± 8.8 
Protein Ankyrin = 18 Ankyrin = 11 Ankyrin = 7 NS 
Band 3 = 1 Band 3 = 0 Band 3 = 1 
Protein 4.2 = 1 Protein 4.2 = 0 Protein 4.2 = 1 
α-Spectrin = 2 α-Spectrin = 2 α-Spectrin = 0 
β-Spectrin = 12 β-Spectrin = 7 β-Spectrin = 5 
Mean hemoglobin, g/L 97 ± 2.2 83 ± 1.9 110 ± 2.6 **** 
Mean reticulocytes, G/L 392 ± 18 391 ± 26 393 ± 24 NS 
Mean total bilirubin, µmol/L 59 ± 6 47 ± 6.9 68 ± 9 NS 
Mean no. of transfusion per year 1.6 ± 0.4 3.1 ± 0.7 0.13 ± 0.02 **** 
Lithiasis, % 32.1 20.0 47.5 ** 
Mean age at surgery, y 7.4 ± 0.6 4.3 ± 0.6 11 ± 0.7 **** 
Mean time of follow-up, y 11 ± 0.7 12 ± 0.9 9.3 ± 1.0 NS 

F, female; g/L, grams per liter; G/L, giga per liter; M, male; NS, not significant.

*P < .05; **P < .01; ****P < .0001.

We confirmed the good hematologic response after STS as previously reported by our group12,13  and others14  (Figure 1A). All patients but 1 experienced a significant increase in Hb level 1 year after surgery. The only primary failure was related to an insufficient removal of splenic tissue (postoperative remnant size: 208% of the volume of a normal spleen). The mean Hb level did not decrease over time, with a slight increase after 10 years of follow-up. At the last end point, a sustainable response (no symptom related to anemia and Hb level >95 g/L) was observed in 69 patients (87%). Fifty-nine patients had received at least 1 transfusion before STS. After STS, 65 patients (82%) remained free of any transfusion. In group A, the transfusion rate decreased from 3.13 to 0.19 transfusions per year per patient (P < .0001). However, as previously reported,12,13  although STS decreased the hemolytic rate, it did not abrogate it. The bilirubin level remained unchanged, and among the 46 patients who did not undergo cholecystectomy at STS, 16 (35%) developed cholelithiasis during the follow-up. Fourteen underwent cholecystectomy with a mean delay of 7.5 years after STS. Gallstones were slightly more frequent in group A than in group B (43% vs 15%; P = .04).

Figure 1

Long-term outcome of STS in HS. (A) Hematologic response after STS: a significant increase in the Hb level (97 ± 1.9 grams per liter [g/L] SEM to 120 ± 1.8 g/L SEM; P < .0001) and a decrease in the reticulocyte count (392 ± 18 giga per liter [G/L] SEM to 291 ± 20 G/L SEM; P < .001) were observed 1 year after STS. These results were stable over time with a mean Hb level being even higher after 10 years (n = 23, P = .04). (B) Evolution of spleen volume after STS assessed by scintigraphy and/or ultrasonography providing 3-dimensional measurement. The remnant was considered as an ellipsoid using the following formula (length × width × thickness × 0.523). Measures were reported to the expected splenic volume for the weight.18-21  Following an 88% reduction after STS (513 ± 60% SEM to 64 ± 5% SEM [n = 40], ****P < .0001), the remnant grew significantly at 1 year (180 ± 12% SEM [n = 50], ****P < .0001), continued to grow at a reduced rate for 5 years after STS (274 ± 29% SEM [n = 21], ***P < .001), and then remained stable (at 10 years: 285 ± 34% SEM [n = 29], not significant [NS]). (C) Spleen volume at 1 year following STS showing significantly faster splenic growth in patients with hematologic relapse (Hb level under 95 g/L) compared with patients with a sustainable hematologic response: median spleen volume at 1 year post-STS: 258 ± 33% SEM (n = 7) vs 165 ± 12% SEM (n = 43), **P = .008. (D) Time of TS according to clinical indication. TS was required earlier when the indication was a hematologic relapse (transfusion dependency and symptomatic anemia/Hb <95 g/L) than any other indications: mean delay 5.3 ± 1 year SEM vs 11.3 ± 1.6 years SEM, respectively (P = .02).

Figure 1

Long-term outcome of STS in HS. (A) Hematologic response after STS: a significant increase in the Hb level (97 ± 1.9 grams per liter [g/L] SEM to 120 ± 1.8 g/L SEM; P < .0001) and a decrease in the reticulocyte count (392 ± 18 giga per liter [G/L] SEM to 291 ± 20 G/L SEM; P < .001) were observed 1 year after STS. These results were stable over time with a mean Hb level being even higher after 10 years (n = 23, P = .04). (B) Evolution of spleen volume after STS assessed by scintigraphy and/or ultrasonography providing 3-dimensional measurement. The remnant was considered as an ellipsoid using the following formula (length × width × thickness × 0.523). Measures were reported to the expected splenic volume for the weight.18-21  Following an 88% reduction after STS (513 ± 60% SEM to 64 ± 5% SEM [n = 40], ****P < .0001), the remnant grew significantly at 1 year (180 ± 12% SEM [n = 50], ****P < .0001), continued to grow at a reduced rate for 5 years after STS (274 ± 29% SEM [n = 21], ***P < .001), and then remained stable (at 10 years: 285 ± 34% SEM [n = 29], not significant [NS]). (C) Spleen volume at 1 year following STS showing significantly faster splenic growth in patients with hematologic relapse (Hb level under 95 g/L) compared with patients with a sustainable hematologic response: median spleen volume at 1 year post-STS: 258 ± 33% SEM (n = 7) vs 165 ± 12% SEM (n = 43), **P = .008. (D) Time of TS according to clinical indication. TS was required earlier when the indication was a hematologic relapse (transfusion dependency and symptomatic anemia/Hb <95 g/L) than any other indications: mean delay 5.3 ± 1 year SEM vs 11.3 ± 1.6 years SEM, respectively (P = .02).

Another concern is the growth of the remnant, consistently noted in previous STS reports.13,14  The volume of the remnant was assessed using scintigraphy or, when not available, ultrasonography providing 3-dimensional measurements.18-21  Surgery reduced by 88% the initial volume of the spleen. The remnant grew quickly during the first year, then at a slower rate until 5 years after surgery without reaching its presurgical volume (Figure 1B). Spleen growth was more pronounced in group A than in group B: at 5 years, remnant volume was found to be 5.5-fold the immediate postoperative volume in group A and 4.4-fold in group B, in the face of similar immediate postoperative volume in both groups. The growth was faster in patients with a hematologic relapse compared with the others (Figure 1C). For patients with a sustained hematologic response, we confirmed our initial report showing no clear correlation between the size of the remnant and the Hb level at the last follow-up (linear regression, R2 0.025, not significant). Twenty-one patients (27%) underwent TS with a mean delay of 8.4 ± 1.2 years after STS (range, 0-20.1 years). The TS rate was higher in group A than in group B (47% vs 8%, P < .0001). In all cases, TS was performed with no major surgical difficulties. One patient underwent early TS because of an immediate postoperative hemorrhage; 1 other for primary failure. Nine patients, all from group A, underwent TS for hematologic relapse when Hb level fell below 95 g/L, with relapses occurring with a mean delay of 5.3 ± 1 years after STS. The 10 other patients underwent TS for reasons not related to anemia: abdominal pain (n = 2), discomfort related to the remnant size (n = 5), major icterus (n = 2), or wandering spleen (n = 1). For these patients, TS was performed later after STS (11.3 ± 1.6 years; P = .02) (Figure 1D). This probably explains the higher rate of TS in our series in comparison with other reports with shorter follow-up. Of note, in group A, none of the preoperative characteristics (including the Hb levels, the spleen volume, or the transfusion rate) was found to be a reliable predictive parameter for future TS.

In summary, with the intrinsic limitations of a single-center retrospective study, our results clarify the improvement that can be expected from STS in HS on a long-term basis. Benefits of STS were different depending on the disease severity. In children under the age of 6 years with severe anemia and/or transfusion requirement, STS was very efficient and safe because it alleviates the transfusion rate and increases the Hb to a level compatible with normal growth and activity. Of interest, half of these patients will not require TS. However, the other half will require TS, but at an age when it will be much safer. In the case of patients with intermediate HS, splenectomy aims to remove discomfort related to the high hemolytic rate. For them, STS, by preserving the spleen functions, appears to be an interesting alternative; it avoids the distressing decision of TS abrogating hemolysis but carrying the long-term risk of infections and vascular events. However, the price to pay is the persistence of a lower but persistent hemolytic state that exposes patients to its own complications.

Authorship

Acknowledgments: This work was funded by the Recherche et Formation en Hématopathologie Association and by the Réseau Hématologie Picardie Association.

Contribution: G.T. and N.M. initiated the study; T.P., C.G., N.M., G.T., and L.G. designed the study and wrote the paper; T.P. and C.G. were in charge of the infants; G.T., A.T., and L.G. were in charge of the adults; F.G. and G.d.L. were the referent surgeons; V.P. performed ektacytometry; and M.F.-T. performed sodium dodecyl sulfate–polyacrylamide gel electrophoresis.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Loïc Garçon, Service d’Hématologie Biologique, CHU Amiens, 80054 Amiens Cedex 1, France; e-mail: garconloic@gmail.com.

References

1
Delaunay
 
J
The molecular basis of hereditary red cell membrane disorders.
Blood Rev
2007
, vol. 
21
 
1
(pg. 
1
-
20
)
2
Molnar
 
Z
Rappaport
 
H
Fine structure of the red pulp of the spleen in hereditary spherocytosis.
Blood
1972
, vol. 
39
 
1
(pg. 
81
-
98
)
3
Cynober
 
T
Mohandas
 
N
Tchernia
 
G
Red cell abnormalities in hereditary spherocytosis: relevance to diagnosis and understanding of the variable expression of clinical severity.
J Lab Clin Med
1996
, vol. 
128
 
3
(pg. 
259
-
269
)
4
Da Costa
 
L
Mohandas
 
N
Sorette
 
M
Grange
 
MJ
Tchernia
 
G
Cynober
 
T
Temporal differences in membrane loss lead to distinct reticulocyte features in hereditary spherocytosis and in immune hemolytic anemia.
Blood
2001
, vol. 
98
 
10
(pg. 
2894
-
2899
)
5
Bolton-Maggs
 
PH
Stevens
 
RF
Dodd
 
NJ
Lamont
 
G
Tittensor
 
P
King
 
MJ
General Haematology Task Force of the British Committee for Standards in Haematology
Guidelines for the diagnosis and management of hereditary spherocytosis.
Br J Haematol
2004
, vol. 
126
 
4
(pg. 
455
-
474
)
6
Bolton-Maggs
 
PH
Langer
 
JC
Iolascon
 
A
Tittensor
 
P
King
 
MJ
General Haematology Task Force of the British Committee for Standards in Haematology
Guidelines for the diagnosis and management of hereditary spherocytosis--2011 update.
Br J Haematol
2012
, vol. 
156
 
1
(pg. 
37
-
49
)
7
Perrotta
 
S
Gallagher
 
PG
Mohandas
 
N
Hereditary spherocytosis.
Lancet
2008
, vol. 
372
 
9647
(pg. 
1411
-
1426
)
8
Guitton
 
C
Garçon
 
L
Cynober
 
T
, et al. 
Hereditary spherocytosis: guidelines for the diagnosis and management in children [in French].
Arch Pédiatr
2008
, vol. 
15
 
9
(pg. 
1464
-
1473
)
9
Bisharat
 
N
Omari
 
H
Lavi
 
I
Raz
 
R
Risk of infection and death among post-splenectomy patients.
J Infect
2001
, vol. 
43
 
3
(pg. 
182
-
186
)
10
Schilling
 
RF
Gangnon
 
RE
Traver
 
MI
Delayed adverse vascular events after splenectomy in hereditary spherocytosis.
J Thromb Haemost
2008
, vol. 
6
 
8
(pg. 
1289
-
1295
)
11
Schilling
 
RF
Spherocytosis, splenectomy, strokes, and heat attacks.
Lancet
1997
, vol. 
350
 
9092
(pg. 
1677
-
1678
)
12
Tchernia
 
G
Gauthier
 
F
Mielot
 
F
, et al. 
Initial assessment of the beneficial effect of partial splenectomy in hereditary spherocytosis.
Blood
1993
, vol. 
81
 
8
(pg. 
2014
-
2020
)
13
Bader-Meunier
 
B
Gauthier
 
F
Archambaud
 
F
, et al. 
Long-term evaluation of the beneficial effect of subtotal splenectomy for management of hereditary spherocytosis.
Blood
2001
, vol. 
97
 
2
(pg. 
399
-
403
)
14
Buesing
 
KL
Tracy
 
ET
Kiernan
 
C
, et al. 
Partial splenectomy for hereditary spherocytosis: a multi-institutional review.
J Pediatr Surg
2011
, vol. 
46
 
1
(pg. 
178
-
183
)
15
Rice
 
HE
Oldham
 
KT
Hillery
 
CA
Skinner
 
MA
O’Hara
 
SM
Ware
 
RE
Clinical and hematologic benefits of partial splenectomy for congenital hemolytic anemias in children.
Ann Surg
2003
, vol. 
237
 
2
(pg. 
281
-
288
)
16
Rice
 
HE
Englum
 
BR
Rothman
 
J
, et al. 
Splenectomy in Congenital Hemolytic Anemia (SICHA) Consortium
Clinical outcomes of splenectomy in children: report of the splenectomy in congenital hemolytic anemia registry.
Am J Hematol
2015
, vol. 
90
 
3
(pg. 
187
-
192
)
17
de Porto
 
AP
Lammers
 
AJ
Bennink
 
RJ
ten Berge
 
IJ
Speelman
 
P
Hoekstra
 
JB
Assessment of splenic function.
Eur J Clin Microbiol Infect Dis
2010
, vol. 
29
 
12
(pg. 
1465
-
1473
)
18
Schlesinger
 
AE
Edgar
 
KA
Boxer
 
LA
Volume of the spleen in children as measured on CT scans: normal standards as a function of body weight.
AJR Am J Roentgenol
1993
, vol. 
160
 
5
(pg. 
1107
-
1109
)
19
De Odorico
 
I
Spaulding
 
KA
Pretorius
 
DH
Lev-Toaff
 
AS
Bailey
 
TB
Nelson
 
TR
Normal splenic volumes estimated using three-dimensional ultrasonography.
J Ultrasound Med
1999
, vol. 
18
 
3
(pg. 
231
-
236
)
20
Hollingsworth
 
CL
Rice
 
HE
Hereditary spherocytosis and partial splenectomy in children: review of surgical technique and the role of imaging.
Pediatr Radiol
2010
, vol. 
40
 
7
(pg. 
1177
-
1183
)
21
Rosenberg
 
HK
Markowitz
 
RI
Kolberg
 
H
Park
 
C
Hubbard
 
A
Bellah
 
RD
Normal splenic size in infants and children: sonographic measurements.
AJR Am J Roentgenol
1991
, vol. 
157
 
1
(pg. 
119
-
121
)

Author notes

*

T.P. and C.G. contributed equally to this work.