Conservative treatment of children with traumatic ruptures of the spleen: results of 22 years of experience

Full Text

Background

Total splenectomy causes the syndrome of post-splenectomy hyposplenism associated with immunodeficiency and hematologic disorders. Fulminant sepsis is the most dangerous late complication after splenectomy with a mortality rate of 50%–70% [1, 2]. Primary prevention of post-splenectomy complications includes spleen preservation through non-surgical management [3, 4]. Non-surgical management of children with ruptured spleen was implemented in our clinic in March 2002 after 5 years of experience in therapeutic and diagnostic laparoscopy. The use of active laparoscopy for splenic injuries in children has shown that laparotomy can be avoided in more than 70% of cases due to the absence of persistent bleeding [5]. The fact that splenic hemorrhage in children resolved spontaneously led to a revision of the treatment strategy in favor of conservative, organ-sparing options. Our study provides a detailed analysis of clinical cases, observational findings, and long-term outcomes, as well as practical recommendations for healthcare professionals.

The study aimed to summarize 22 years of experience in the management of children with traumatic splenic rupture.

Materials and methods

This observational single-center prospective study was conducted from March 2002 to March 2024 at the Ivano-Matreninskaya City Children’s Clinical Hospital in Irkutsk, Russia. Data of 95 children with traumatic splenic rupture were evaluated. Non-surgical treatment was received by 83 (87.4%) patients and surgical treatment was received by 12 (12.6%) patients. The age of the affected children was 12 [8; 14] years, with 3.3 times more boys observed (73 vs. 22).

Eligibility criteria. Inclusion criteria were blunt abdominal trauma, splenic rupture, hemoperitoneum. Non-inclusion criteria were splenic injury without peritoneal bleeding, concomitant abdominal trauma with hemoperitoneum.

Intervention description

When a child presented to an Emergency Unit with abdominal pain and a history of abdominal trauma, a physical examination and hemodynamic assessment were performed. The Allgower’s Shock Index (SI) was calculated as the ratio of maximum heart rate (HR) to minimum systolic blood pressure (SBP). The shock index was adjusted for childhood age. SI cut-off values were >1.2 (age up to 6 years), >1.0 (7–12 years), and >0.9 (over 12 years) [6]. An elevated SI was defined as a criterion for hemodynamic instability in children. The hemodynamic status was considered as a factor of the treatment choice in patients with splenic rupture.

If hemodynamics were stable and abdominal ultrasound showed hemoperitoneum, the patient was admitted to a Surgery Unit with physical activity restriction for 3 days. Physiological functions (SBP, heart rate, SI, urine output, temperature) were monitored every 6 hours. Laboratory tests (hemoglobin, platelet count, white blood cell count, erythrocyte sedimentation rate) were evaluated. Hemostatic and antibacterial therapy was prescribed for 3–7 days. Echo ultrasound was performed every 6 hours for the first day, then on the second day, and then at discharge.

If the SI was elevated, the child was admitted to the Intensive Care Unit (ICU). Bolus therapy was administered at 20–40 mL/kg of body weight. Clinical and laboratory parameters were monitored. If hypotension persisted despite the use of crystalloids, blood components were transfused at 10–15 mL/kg body weight. If the hemoglobin level was below 70 g/L, blood transfusion was also performed. If the SI remained elevated despite blood transfusion, an emergency laparotomy was performed. To monitor peritoneal blood volume, ultrasound monitoring was performed every 2 hours for 6 hours after admission, then every 6 hours for the first day and once a day thereafter [7–9].

To calculate the abdominal volume, the method for determining the free fluid volume in the abdomen (patent RU No. 2830196 C1, issued on November 14, 2024) was used;

V = V_ultrasound × BSA × К;

where, V is the hemoperitoneum volume; V_ultrasound is the hemoperitoneum volume calculated using the formula for an ellipsoid; BSA is the body surface area, m²; К is a coefficient equal to 2.35 for hemoperitoneum up to 11 mL/kg of body weight, 1.35 for hemoperitoneum 11–22 mL/kg of body weight and 1.0 for hemoperitoneum more than 22 mL/kg of body weight.

All children were followed for complications related to spleen injury. Comprehensive follow-up of patients after discharge ranged from 6 months to 15 years.

Subgroup analysis. For analysis, patients were divided into two groups: the control group (n = 62; 65.3%) in the early treatment period (March 2002 to August 2012); the study group (n = 33; 34.7%) in the late treatment period (September 2012 to March 2024).

Statistical analysis. Data samples were tested for normality of distribution at a significance level of p < 0.05 (frequency histogram, Lilliefors test, Shapiro-Wilk test). The significance of differences between groups was assessed using a non-parametric procedure; results were presented as median (Me), 25th and 75th quartiles [Q₁; Q₂]. Groups were compared using the Mann–Whitney U test Categorical criteria were compared using the chi-squared test (χ²). Discriminant analysis was used to predict outcomes when assessing the combined differences in multiple factors between groups. The risk prediction model was represented graphically by constructing a receiver operating characteristic (ROC) curve, which is a graphical representation of the dependence of two values: sensitivity (Se) and specificity (Sp). The corresponding area under the curve (AUC) was calculated. The model was tested for validity using the Hosmer–Lemeshow test, and the null hypothesis was accepted if p > 0.05. Statistical data were processed using Statistica v.10.1, MedCalc statistical software. The confidence level was set at p < 0.05.

Results

Injury-to-hospitalization time was 2.9 [1.6; 7.1] hours; 73.7% of children were admitted within the first 6 hours. Of 95 patients, 62 (65.3%) children required hospitalization in the ICU. The number of bed days was 3.5 [2; 5] days in the ICU and 8.7 [6; 13] days in the Surgery unit. The spring/summer period was the season with the highest hospitalization rate at 67.4% of the total patient cohort.

The surgical treatment rate was 12.6% (n = 12). Initially, four (4.2%) patients received surgical treatment within the first 2 hours of admission; three patients underwent laparotomy due to hemodynamic instability; one patient underwent laparoscopic exploration due to risk of the hollow organ injury. Due to persistent bleeding, six patients (6.3%) underwent surgery, five patients after 6–12 hours of hospitalization and one patient after 24 hours of hospitalization. Persistent intraperitoneal bleeding with a tendency to hypotension was an indication for surgery in 4 patients. In 2 cases, laparoscopic abdominal lavage was performed at 12 hours after admission due to persistent abdominal pain and unclear ultrasound and radiologic findings. Two patients (2.1%) underwent laparotomy on days 5 and 6 after injury due to delayed bleeding.

Of 95 patients, 2 (2.1%) underwent splenorrhaphy, 3 (3.1%) underwent laparoscopic exploration of the spleen, and 7 (7.4%) underwent splenectomy. All splenectomized patients underwent autotransplantation of splenic tissue. In he surgical management group, children required long-term combination antibacterial therapy for 13 [10; 16] days, which was statistically significantly longer than in the non-surgical management group, where antibacterial prophylaxis was given for 5 [3; 7] days (p = 0.001). If the spleen was removed after discharge from the hospital, patients were prescribed preventive vaccinations.

Discriminant analysis was used to identify the combined variables associated with surgical treatment (Table 1).

 

Table 1. Discriminant analysis of factors influencing the choice of surgical treatment for splenic rupture in children (n = 71)

Таблица 1. Дискриминантный анализ факторов, влияющих на выбор хирургического лечения при разрыве селезенки у детей (n = 71)

Factor	Variables
	Me [25; 75]	Wilks’ Lambda	Partial Lambda	F-remove (1,58)	p-value	1-Toler. (R-Sqr.)
Age, years	13 [7; 14]	0.417	0.997	0.186	0.667	0.232
Systolic blood pressure, mmHg	95 [70; 118]	0.489	0.848	10.35	0.002	0.936
Diastolic blood pressure, mmHg	55 [40; 60]	0.417	0.995	0.287	0.594	0.871
Heart rate, per minute	105 [100; 120]	0.457	0.910	5.762	0.019	0.961
Glasgow Coma Scale	15 [13; 15]	0.416	0.999	0.042	0.839	0.291
Vegetative Kerdo’s index, relative units	48 [39; 60]	0.415	0.999	0.001	0.979	0.918
Shock index	1.1 [0.9; 1.57]	0.529	0.784	15.95	0.001	0.971
Blood loss at admission, %	13 [6.3; 19]	0.516	0.805	14.06	0.001	0.828
Maximum blood loss	2 [1; 3]	0.600	0.692	25.81	0.001	0.827
Injury-to-hospitalization time, hours	2 [1; 6]	0.424	0.9792	1.229	0.272	0.158

Note. Result of discriminant function analysis. Number of variables in the model: 10; grouping: by surgery. Wilks’ Lambda: 0.415, approx. F(10,58) = 8.16, p < 0.0000

Примечание. Итог анализа дискриминантной функции. Количество переменных в модели: 10; группировка: операция. Wilks’ Lambda: 0,415 approx. F(10,58) = 8,16, p < 0,0000

 

Statistically significant combinations of factors of the surgical treatment choice for traumatic splenic rupture were identified: low SBP of 95 [70; 118] mmHg, p = 0.002; tachycardia with HR of 105 [100; 120] per minute, p = 0.019; elevated SI of 1.1 (0.9; 1.57), p = 0.001; blood loss at admission of 13% [6.3; 19] of circulatory volume, p = 0.001; maximum blood loss of 2 [1; 3], p = 0.001.

The ROC curve in Figure 1 shows the overall reliability of the model for the identified factors associated with surgical treatment.

 

Fig. 1. Receiver operating characteristics (ROC) curve based on combined factors: systolic blood pressure (1), heart rate (2), shock index (3), % percentage of blood loss from circulating blood volume (4), and maximum degree of blood loss (5)

Рис. 1. Кривая рабочих характеристик по совокупности факторов систолического артериального давления (1), частоты сердечных сокращений (2), индекса Альговера (3), % кровопотери от объема циркулирующей крови (4) и максимальной степени кровопотери (5)

 

The model validity test was positive (χ² = 12.3; p = 0.14). The test sensitivity was 95.2% [95% confidence interval (CI): 86.7–99.0], specificity was 100% (95% CI: 59–100). The overall confidence level was 95.7% (95% CI: 88.0–99.1; Table 2).

 

Table 2. Predictive accuracy of ROC curve for identified factors in discriminant analysis

Таблица 2. Прогностическая ценность ROC-кривой выявленных факторов при дискриминантном анализе

Parameters	AUC	95% confidence interval
Systolic blood pressure	0.68 ± 0.13	0.56–0.79
Heart rate	0.70 ± 0.08	0.58–0.81
Shock index	0.77 ± 0.08	0.66–0.87
Blood loss at admission, %	0.69 ± 0.07	0.57–0.79
Maximum blood loss	0.74 ± 0.09	0.62–0.84
Combination of factors	0.90 ± 0.07	0.81–0.96

 

The model fit each identified factor well (AUC > 0.6). When the identified parameters were pooled, the area under the curve (AUC) was 0.90 ± 0.07 (95% CI: 0.81–0.96), indicating excellent model quality.

Five combined factors of the surgical treatment choice were identified, which were associated with blood loss and hemodynamics.

In the early stages of management of splenic rupture in children, the factors of the surgical treatment choice included polytrauma, persistent bleeding, and depression of consciousness. The combination of blood loss and hemodynamic factors has been used since September 2012 to divide patients into two groups. It should be noted that the incidence of splenic injury has decreased 1.9-fold over the past decade (62 vs. 33).

Table 3 shows clinical and epidemiological characteristics of patients in control and study groups.

 

Table 3. Characteristics of patients with splenic rupture in early and late treatment groups

Таблица 3. Характеристики пациентов с разрывом селезенки в группах раннего и позднего периодов лечения

Parameters	Control group (n = 62), Me [Q1; Q2]	Study group (n = 33), Me [Q1; Q2]	p
Age, years	11 [8; 14]	12 [7; 14]	0.75
Injury-to-hospitalization time, hours	2.3 [1; 8]	3.5 [2; 7]	0.2
Number of bed days in the Intensive Care Unit	3 [2; 4]	2.8 [2; 5]	0.4
Number of bed days in the Surgery Unit	12 [8; 14]	7 [7; 9]	0.001
Injury Severity Score (ISS)	9 [9; 13]	9 [9; 13]	0.9
Glasgow Coma Scale, score	15 [15; 15]	15 [15; 15]	0.7
Ultrasound volume of hemoperitoneum at admission, mL	200 [100; 400]	175 [90; 350]	0.45
Ratio of hemoperitoneum volume to admission weight, mL/kg	4.9 [2.4; 9.1]	4.5 [2.7; 8.1]	0.54
Blood loss at admission, % of circulatory volume	7.1 [3.4; 13]	6.4 [3.8; 11.6]	0.54
Maximum blood loss, % of circulatory volume	8.5 [4; 17.6]	11.4 [5.5; 16.2]	0.72
Maximum blood loss	1 [1; 2]	1 [1; 2]	0.81
Systolic blood pressure, mmHg	110 [100; 118]	115 [106; 120]	0.11
Diastolic blood pressure, mmHg	65 [60; 77]	70 [63; 75]	0.13
Heart rate, per minute	100 [90; 110]	98 [90; 110]	0.51
Shock index	0.9 [0.8; 1.1]	0.85 [0.76; 1.0]	0.27

 

No differences were found in age, injury-to-hospitalization time, or bed days in the ICU. A statistically significant difference was found in the number of days spent in the Surgery Unit: the duration of hospital stay was 12 [8; 14] days in the control group and 7 (7; 9) days in the study group (p = 0.001). Patients in the groups compared did not differ in ISS, Glasgow Coma Scale, blood loss, or hemodynamics. Table 4 shows the main categorical parameters of the treatment choice in the evaluated groups.

 

Table 4. Key categorical parameters in groups of different observation periods Таблица 4. Основные категориальные параметры в группах разных периодов наблюдения

Parameters	Control group (n = 62)		Study group (n = 33)		p
	n	%	n	%
Polytrauma Yes No	23 39	37.1 62.9	10 23	30.3 69.7	0.51
Persistant bleeding: Yes No	14 48	22.6 77.4	10 23	30.3 69.7	0.41
Elevated Allgower’s Shock Index at admission: Yes No	16 46	25.8 74.2	7 26	21.2 78.8	0.62
Blood transfusion: Yes No	7 55	11.3 88.7	5 28	15.2 84.8	0.59

 

Over 22 years, there were no statistically significant differences in the incidence of polytrauma, persistent bleeding, hemodynamic parameters, or blood transfusions between the groups. However, it should be noted that over the past 11 years, the incidence of splenic rupture surgery decreased 2.6-fold from 16.1% to 6.1%. Non-surgical management of splenic rupture in children in the study group was 93.9%.

When evaluating immediate outcomes after splenectomy, it was found that 71.4% (n = 5) of the children had thrombocytosis on day 3–6 after surgery; the platelet count was 430 [369.8; 431] × 10⁹/L. In the non-surgical management group, the maximum platelet count during the 3–6 day period was 256 [210; 279] × 10⁹/L (p = 0.002). After spleen removal, all children had an elevated ESR of 25 [23; 39] mm/h for 2 weeks. In the non-surgical management group, the ESR on days 5–7 was 6 [4; 8] mm/h (p = 0.001).

To evaluate the long-term outcome of splenic rupture in children, 100% (n = 95) of patients were followed up for the first 6 months, 38.9% (n = 37) were followed up between 6 months and 3 years, and 10.5% (n = 10) were followed up at 3 years and later. In the late phase after splenectomy, three patients (42.8%) had frequent infections up to 5–7 times per year. One patient (14.3%) had up to 3–4 infections per year. No hyposplenism was reported during the non-surgical management of splenic rupture.

Discussion

Non-surgical management of splenic rupture in children is becoming the gold standard of care. Preservation of the injured spleen is the goal of primary prevention of hyposplenism and its complications associated with post-splenectomy infection [10]. International guidelines for non-surgical management of children with splenic rupture have been published in the last decade [11–13]. Foreign literature showed that when these guidelines were followed, the rate of non-surgical management was 42.5%–97.2%, while the rate of surgical treatment varied from 2.8% to 31.8%, and splenectomy was performed in 0.0%–7.5% of children. Special focus should be paid to the widespread use of angioembolization in children with splenic injury, ranging from 1.2% to 42.5% according to the references reviewed [14–17].

The literature review highlights papers that address the practice of treating children with splenic injury. For example, Filipescu et al. [18] showed that the incidence of splenectomy during treatment was 1.5% in a pediatric hospital and 14.4% in an adult hospital (p = 0.01). The authors reported that children admitted to pediatric hospitals had a higher severity of splenic injury. However, children were more likely to receive blood transfusions in adult hospitals. Miyata et al. [19] showed that in clinics with extensive experience in the management of children with splenic rupture, the rate of splenectomy was low (odds ratio: 0.5; 95% CI: 0.3–0.8). Gorelik et al. [20] showed that the effectiveness of conservative management of traumatic splenic rupture in children during primary hospitalization in a tertiary hospital was 94.1%. In community hospitals, 58.8% (n = 53) received surgical treatment, with the rate of splenectomy of 47.8% (n = 43). The study data were consistent with our findings that the incidence of surgery decreased from 16.1% to 6.1% with long-term patient management.

In Russian publications, the rate of laparoscopy for splenic injury with hemoperitoneum remains high [21, 22]. In our study, 3 (3.1%) children required laparoscopic treatment. The following practically significant facts were highlighted in the literature, which led to the decision to stop the routine use of laparoscopic treatment of splenic injury in children: the incidence of spontaneous hemostasis of splenic rupture in children was 78%–85%, laparoscopy for isolated splenic injury was a diagnostic sign in only 61%–72% [23–26]. The decision was made to use minimally invasive treatment options in patients with persistent abdominal pain and controversial findings of ultrasound and multispectral computed tomography. With the parameters described, laparoscopy is very rarely used for abdominal injuries. In the American College of Surgeons Pediatric Trauma Centers Consortium (ATOMAC) study, 5 (1.3%) of 410 children with splenic injury underwent laparoscopy 42 [21; 90] hours after admission. Laparoscopy was usually required for concomitant abdominal injury [26].

When the immediate outcomes of splenic rupture treatment were evaluated, it was found that splenectomy was associated with thrombocytosis and elevated ESR. A literature review found splenectomy in 4.5% of all causes of thrombocytosis in children [27]. Thrombocytosis can lead to thrombosis due to blood clotting disorders in blood vessels. Thrombosis of the portal vein system is particularly dangerous in the early post-splenectomy period and can develop for up to 1 month after surgery [28]. Deep vein thrombosis and pulmonary embolism develop in up to 3.5% of adolescents aged 12–17 years in the late post-splenectomy period [29]. A persistent increase in body temperature and persistently elevated CRP levels (up to 55 mg/dL), leukocytosis (up to 22 × 10⁹/L), and elevated ESR (up to 20 m/h) were reported in children during the first week after splenectomy [30].

In our study of long-term splenectomy outcomes, 57.1% of patients reported frequent infections. The data obtained are consistent with other scientific papers. Splenectomy has been shown to significantly disrupt the interaction between T and B lymphocytes, leading to antiviral resistance. Over 90% of children are predisposed to frequent seasonal viral infections, herpes, inflammatory skin lesions, and exacerbation of chronic diseases [30, 31].

Non-surgical management of children with splenic rupture is shown not to cause symptoms of hyposplenism.

Conclusion

Twenty-two years of treating children with traumatic splenic rupture have demonstrated the clinical efficacy and safety of non-surgical management which is possible in 93.9% of cases. Many years of our experience confirm the effectiveness of the used organ-preserving options. These findings confirm the importance of a personalized approach for each pediatric patient based on both the clinical picture and the regenerative potential of the body. The non-surgical option not only minimized the need for surgery, but also helped to better restore spleen function, which is a way of primary prevention of hyposplenism.

Therefore, our results serve as a basis for further research in this area and highlight the need to develop a strategy to preserve the spleen that will undoubtedly have a positive impact on the long-term health of children who have experienced traumatic injury to this organ.

Additional info

Authors’ contribution. All authors made a substantial contribution to the conception of the study, acquisition, analysis, interpretation of data for the work, drafting and revising the article, final approval of the version to be published and agree to be accountable for all aspects of the study. The contribution of each author: V.V. Podkamenev — the concept and design of the study, writing a text, editing; I.A. Pikalo — the concept and design of the study, collection and processing of material, patient supervision, statistical analysis, writing a text, editing; V.A. Novozhilov, N.I. Mikhailov, E.M. Petrov, V.Kh. Latypov, S.V. Moroz, D.Yu. Khaltanova — patient supervision, forming of a scientific database and its analysis; O.A. Karabinskaya — processing of material, statistical analysis.

Funding source. This study was not supported by any external sources of funding.

Competing interests. The authors declare that they have no competing interests.

Ethics approval. The study was approved by the local ethical committee of the Irkutsk State Medical University (protocol No. 3 dated 2019 Nov 15).

About the authors

Vladimir V. Podkamenev

Irkutsk State Medical University

Email: vpodkamenev@mail.ru
ORCID iD: 0000-0003-0885-0563
SPIN-code: 7722-5010

MD, Dr. Sci. (Medicine), Professor

Россия, Irkutsk

Ilia A. Pikalo

Irkutsk State Medical University

Author for correspondence.
Email: pikalodoc@mail.ru
ORCID iD: 0000-0002-2494-2735
SPIN-code: 4885-4209

MD, Cand. Sci. (Medicine)

Россия, Irkutsk

Vladimir A. Novozhilov

Irkutsk State Medical University; Ivano-Matreninskaya City Children Clinical Hospital

Email: novozilov@mail.ru
ORCID iD: 0000-0002-9309-6691
SPIN-code: 5633-5491

MD, Dr. Sci. (Medicine), Professor

Россия, Irkutsk; Irkutsk

Olga A. Karabinskaya

Irkutsk State Medical University

Email: fastmail164@gmail.com
ORCID iD: 0000-0002-0080-1292
SPIN-code: 1511-3402

MD, Cand. Sci. (Medicine)

Россия, Irkutsk

Nikolai I. Mikhailov

Ivano-Matreninskaya City Children Clinical Hospital

Email: mni.irk@ya.ru
ORCID iD: 0000-0002-7428-3520
SPIN-code: 1153-3175

MD, Cand. Sci. (Med.)

Россия, Irkutsk

Evgenii M. Petrov

Ivano-Matreninskaya City Children Clinical Hospital

Email: emp1976@rambler.ru
ORCID iD: 0000-0002-1083-0951
SPIN-code: 9949-7707
Россия, Irkutsk

Vyacheslav Kh. Latypov

Ivano-Matreninskaya City Children Clinical Hospital

Email: slavalat@gmail.com
ORCID iD: 0009-0005-9147-3309
Россия, Irkutsk

Sergey V. Moroz

Ivano-Matreninskaya City Children Clinical Hospital

Email: moroszsv@mail.ru
ORCID iD: 0009-0002-1202-1127
SPIN-code: 4915-5348
Россия, Irkutsk

Dora Yu. Khaltanova

Ivano-Matreninskaya City Children Clinical Hospital

Email: khaltanovad@mail.ru
ORCID iD: 0000-0001-7018-3007
SPIN-code: 8185-7522
Россия, Irkutsk

References

Supplementary files