Computed Tomography Grade of Splenic Injury Is Predictive of the Time Required for Radiographic Healing

By James M. Lynch, Manuel i? Meza, Beverly Newman, Mary J. Gardner, and Craig T. Aibanese Pittsburgh, Pennsylvania

0 It is largely unknown when a child who has suffered a splenic laceration can return to full unrestricted activity. The purpose of this prospective study is to establish whether the grade of splenic injury is predictive of the length of time required for radiographic healing, and to determine whether there are any adverse long-term sequelae after resumption of unlimited activity. Sixty-nine patients underwent success- ful nonoperative management (NOM) of computed tomogra- phy (CT)-documented splenic injury over a 4-year period. Fifty-eight patients completed follow-up. Mean age was 9.8 years (range, 1 to 17) and mean injury severity score (ISS) was 14.4 (range, 4 to 38). Mechanisms of injury were motor vehicle accident (n = II), motor vehicle pedestrian (n = 51, falls (n = 131, bike crashes (n = 121, sports (n = 81, all-terrain vehicle (n = 4), and horse (n = 5). The CT-documented injury was identified by discharge ultrasound scan (US) in all cases. There were no long-term complications. Mean time to US healing in grade I (n = 9). II (n = 261, Ill (n = 191, IV (n = 4) injuries was 3.1, 8.2, 12.1, and 20.7 weeks, respectively. P values were significant (P < .Ol) in all cases when compared with the next lower injury grade. The time to radiographic healing is directly proportional to the severity of the splenic injury. There was excellent correlation between the initial CT scan and identification of the injury on the discharge US. No long-term complications leg, delayed splenic rupture, splenic pseudocyst) were seen in this study. Pediatric patients who have suffered splenic injury can safely return to full unre- stricted activity when the US documents healing. Copyright o 1997 by W.B. Saunders Company

INDEX WORDS: Splenic trauma, splenic injury, splenic heal- ing, ultrasonography, splenic injury grading, abdominal in- jury.

N ONOPERATIVE management (NOM) of pediatric blunt splenic trauma is safe and effective.ld6 A

well-accepted principle of NOM is an obligatory period of restricted physical activity after discharge to prevent recurrent bleeding. What has not been defined in the literature is when, and by what criteria, children should be allowed to resume unrestricted activity.

Serial outpatient abdominal ultrasonography (US) may provide objective data to suggest splenic healing based on resumption of normal homogeneity of the organ compared with baseline (postinjury) US examination. If US resolution of the splenic injury is equivalent to true anatomic integrity of the spleen, then it follows that serial US examinations may be used to define the natural history of pediatric splenic injuries managed nonopera- tively.

All children who have suspected blunt splenic injury treated at our institution undergo admission computed tomographic (CT) scanning of the abdomen and pelvis.

JournalofPediatricSurgery, Vol32,No 7 (July),1997: pp 1093-1096

At the time of discharge, and periodically during the period of restricted physical activity, US examinations are performed. The purpose of this study is to determine (1) if the grade of a blunt splenic injury can predict the length of time required for US documentation of injury resolution, and (3) if return to unrestricted physical activity after resolution of the injury by US is safe and sufficient to prevent future complications.

MATERIALS AND METHODS

Children’s Hospital of Pittsburgh (CHP) IS a Level I Regional Resource Pediatric Trauma Center servmg the area of western Pennsyl- vania, northern West Vigmia. and eastern Ohlo. From January. 1992 through December, 1995. all sur\~ivmg patlents with the diagnosis of blunt splenic Injury discharged from the trauma service after NOM were entered mto this prospective study. Patients who had multisystem as well as isolated splemc mjury were included. Our long-standing protocol for the NOM of splemc injuries is as follows. The diagnosis 1s established and the severity of the injury 1s graded by the system described by Buntain using CT scanning’ (Table I ). For this study, all CT scans are revlewed and graded by a single radiolopist (MPM). All hemodynamically stable patients are placed at strict bed rest and are monitored on the pediatric ward except for those who have mulhple system or grade IV injuries, who are admitted to the Intensive care unit. Patients are allowed to ambulate when they are pain free and then- vital signs and hematocrit are stable. To follow resolution of the injury. a baseline abdominal US is obtained on the day before discharge. Follow-up outpatient US and physical exanunation are performed at J- to 6-week Intervals until complete homogeneity of the splenic tissue 1s observed without residual defect or fluid. Patients who have grade I Injuries. however. had US performed earlier (1 to 2 weeks after discharge).

Patients were excluded if a baseline US examination was not obtamed (n = 0), If follow-up US exammations were not obtamed or the patient failed to return for follow-up (n = 9). or If an injury previously noted on CT scan was not found on the mitml US (n = 2). Data collected mcluded age, sex. mechamsm of injury, CT grades. injury severity scores (KS). correlation of US with CT scan findings. time to US resohmon of the splemc abnormality, and complications. Data were analyzed using a standard commercial statIstica program (STAT View 4.5. Abacus Concepts. Berkeley, CA). Analysis of variance (ANOVA) was performed using I test for unpaired data with post hoc

From the Benedum Pediatvlc Tr-aunza Plvgram, the Divisions of Prdiatrlc Surgery arid Peakztnc Radiology Children’s Hosprtul of Plttshurgh, arid the Departments of Scwge~ and Radiology, Univerxt?, of Pittsburgh School of Me&me, Pittsburgh. PA

Presented at the I996 A~mual Me&rng of the Sectron on Surgery of the Ame~warz Acadenql of Pediatrics. Boston, Massachusetts, October 26-30. 1996.

Address reprint requests to James M. Lynch. MD, Dwector of the Benedum Pediatric Trauma Psogram. Lk\wion of Pediatuc Surgeq, Chddren k Hospital of Pittsburgh, 3705 F@h Axse, P~ttsbwgh, PA 15213.

Copyrght 0 1997 by WB. Snuxders Comparz!~ 0022-3468/97/3207-0033$03 00/O

1093

1094

Table 1. Buntain’s Classification of Splenic Injury

LYNCH ET AL

DISCUSSION

The results of this study demonstrate that injury grade of blunt pediatric splenic trauma as determined by CT scan is predictive of the time for US documentation of a normal-appearing spleen, that baseline US result does correlate with that of the initial CT scan, and that patients can be allowed to return to unrestricted activity without complication when the spleen normalizes as determined by US. These data are supported, in part, by three retrospective published studies of blunt pediatric splenic trauma. Bethel et al* used 9gmTc sulfur colloid scans to follow the healing of splenic injuries in 20 patients. Twenty weeks after injury, six patients had normal studies, four had minimal residual defects, and IO had significant improvement. They were unable to draw any conclusions about the severity of the initial injury and the length of time required for healing. Pranikoff et al9 reported on CT-demonstrated healing of IO of 13 grade I and II injuries, but only 1 of 12 grade III or higher injuries 6 weeks after injury. Unfortunately, those without com- plete healing at 6 weeks were not followed up with further imaging to document time to complete healing, especially in the more severely injured spleen. Benya et allo also used CT scan follow-up to demonstrate 100% resolution of 15 grade I and II injuries at 4 months, 10 of 11 grade III injuries by 6 months, and all 11 grade IV injuries by 48 weeks. Obviously the long intervals between injury and follow-up imaging resulted in a very long period of restricted activity for the patients, much of it probably unnecessary.

Unlike the three aforementioned studies using nuclear medicine imaging or CT scanning, we believe that follow-up can be reliable and accurately obtained using US with its inherent benefits compared with the other two modalities (lower expense, no contrast exposure, lack of ionizing radiation). Ultrasound scan has been reliably used in a number of studies to identify and follow-up splenic injuries.“-I3 In only two patients (both with CT grade I injuries) was the initial injury not able to be identified on the basis of baseline US, and thus, both were excluded from the study.

Based on the data in the present study, we are now more selective as to the time to the first outpatient US examination based solely on the CT grade of the injury. We obtain a follow-up ultrasound at 4 weeks for grade I, 8 weeks for grade II, 12 weeks for grade III, and 20 weeks for grade IV lesions providing that their physical exami- nations remain normal. If healing is documented at these visits, unrestricted activity is allowed. If healing is not sufficient, we obtain US at 4 week intervals until healing is complete. To date, we have not experienced any complications over the long term using this protocol.

Grade I Localized capsular disruption or subcapsular

hematoma without significant parenchymal Injury.

Grade II Single or multiple capsular and parenchymal dis-

ruptions that do not extend into hilum or involve major vessels.

Grade Ill Deep fractures, single or multiple, extending into the hllum and involving major blood vessels.

Grade IV Completely shattered or fragmented spleen, or separated from Its normal blood supply at the

pedlcle.

Modified from Buntain et aI,7

testing (Fisher’s PLSD). Probabilities less than .Ol were considered significant.

RESULTS

Sixty-nine patients were originally eligible for entry into the study. In two patients, initial US could not be correlated with the initial CT findings (Grade I injuries), thus they were excluded. Nine patients had incomplete data because of failure to keep trauma clinic appoint- ments or they lived too far away and wished to be followed up by surgeons in their area. Telephone follow- ups were performed in these patients, and no untoward events were reported.

Fifty-eight patients (48 boys and 10 girls) had com- plete data and met all study criteria. Mean age was 9.8 years (range, 1 to 17) and mean ISS was 14.4 (range, 4 to 38). Mechanisms of injury were passenger in motor vehicle crash (n = ll), pedestrian struck by a motor vehicle (n = 5), falls (n = 13), bike crashes (n = 12), sports related (n = 8), all terrain vehicle (n = 4), and horse (n = 5).

The CT scan identified nine grade I, 26 grade II, 19 grade III, and four grade IV splenic injuries (Table 2). US correlated with the initial CT scan in 67 of 69 initial cases and in all study cases. Although an area of abnormality was seen on the US and correlated with the CT, fre- quently, especially in the early postinjury period, the findings were subtle and much more difficult to appreci- ate than on CT.

Table 2 demonstrates that the time to US resolution of the splenic abnormality varied directly with the severity of the CT grade. There were no complications of NOM in these 58 patients. After resuming unrestricted activity, there were no untoward events such as pain, abscess, or rebleeding.

Table 2. lime to Splenic Healing According to CT Injury Grade

Injury grade I II Ill IV

No. of patients 9 26 19 4

Time (wk)* 3.11 i 3.06 8.21 i- 3.86”’ 12.11 i 5.32** 20.71 t 7.72**

*Mean t SD. **P < .Ol when compared with next lower injury grade.

CT GRADE AND SPLENIC HEALING 1095

REFERENCES

1. Bond DJ, Eichelberger MR, Gotschall CS, et al: Nonoperative management of blunt hepatic and splenic inJury in children. Annals of Surg 223:286-289, 1996

2. Coburn MC, Pfeifer .I, DeLuca FG: Nonoperative management of splenic and hepatic trauma in the multiply inJured pediatric and adolescent patient. Arch of Surg 130:332-338. 1995

3. Morse MA, Garcia VF: Selective nonoperative management of pediatric blunt splenic trauma: Risk for missed associated injuries. J Pediatr Surg 29123-27, 1994

4. Tepas JJ 3d: Blunt abdominal trauma in children. Curr Opin Pediatr 5:317-324, 1993

5. Velanovich V, Tapper D: Decision analysis m children with blunt splenic trauma: The effects of observation, splenorrhaphy, or splenec- tomy on quality-adjusted life expectancy. J Pediatr Surg 28:179-185. 1993

6. Pearl RH, Wesson DE. Spence LJ, et al: Splenic inJury: A 5-year update with improved results and changing criteria for conservative management. J Pediatr Surg 24:121-124, 1989

7. Buntain WL, Gould HR. Maul1 Kl: Predictability of splemc salvage by computed tomography. J Trauma 28:24-31. 1988

8. Bethal CA, Toulouktan RJ, Seashore JH. et al: Outcome of nonoperative management of splemc injury with nuclear scanning. Clinical stgnificance of persistent abnormahties. Am J Dis Child 146:198-200, 1992

9. Pranikoff T. Hirsch1 RB, Schlesinger AE. et al: Resolution of splenic injury after nonoperative management. J Pediatr Surg 29:1366- 1369,1994

10. BenyaEC. Bulas DJ. Eichelberger MR. et al: Splemc injury from blunt abdominal trauma m children: Follow-up evaluation with CT. Radiology, 195:685-688, 1995

11. Adler DD, Blane CE, Coran AG, et al: Splenic trauma m the pediatric patient: The integrated roles of ultrasound and computed tomography. Pediatrics 78:576-580, 1986

12. Roche BG. Bugmann R, LeCoultre C: Blunt mjuries to hver. spleen, kidney and pancreas in pediatric patients. Eur J Pediatr Surg 2:154-156, 1992

13. Siniluoto TM. Paivansalo MJ, Lanmng FR, et al: Uhrasonogra- phy in traumatic splenic rupture. Clm Radio1 46:391-396, 1992

Discussion

D. Lund (Boston, MA): In Boston, we admit 30 kids a year with ruptured spleens, and it is unclear to me how long we should be keeping these kids on restricted activity.

Do you think that the difference in healing rates that you see has to do with the fact that the CT scan is clearly more sensitive than the ultrasound? The radiology litera- ture shows that ultrasound can miss anywhere from 20% to 80% of the initial spleen injuries, so I wonder if that is playing a role.

J.M. Lynch (response): Our radiographers can tell us down to about 2 mm if there is a defect in the spleen, and that is why we wanted to be sure we could see the defect on discharge.

One of the problems though, is that the defects are much harder to see earlier than later on because there is not a good definition between the fluid component and the solid component of the spleen.

There were some patients in the study who on their third or fourth follow-up had a defect of 3 mm. Just to be true to the study we did not call those healed, we brought them back, and continued to follow them up.

We had two patients from whom I obtained follow-up CT scans because the healing did not appear to be going the way we liked it. The CT scan results correlated quite well with the size of the defect the ultrasonographers saw.

R. Pearl (Toronto, Ontario): Much has changed in the nonoperative therapy of ruptured spleens over the last 20 years. At our center no one goes to the intensive care unit anymore. Everybody stays in the hospital for about 5 to 7

days. We discharge them, as you do, when they are well and asymptomatic.

The problem I have with your discussion, though, is I am not sure what radiographic healing means. I will give you an example. We have had a couple of kids who healed with spleens in two pieces, permanently. The spleens were fractured down the middle. When they healed you can get a nice picture of a completely heeled spleen in two pieces.

So radiographic healing might show an anatomical repair but I think the physiological repair and the safety for the child to perform sports is not correlated.

J.M. Lynch (response): That is one of the reasons why we titled the paper the way we did, because we realized we were defining radiographic healing. We have followed up these patients for a month or two after stopping the ultrasound scan and sent them back on unlimited activity. They have had no problem since that time. So for our follow-up period we haven’t had any problems.

I mentioned earlier that we did obtain a CT scan on two patients this year. And that was precisely the reason we did. They had healed spleens but they had a defect in the middle because their spleens were cracked in half.

R. Pearl (Toronto, Ontario): In the last 28 years we have not had one child who had a ruptured spleen come back with a secondary injury to their spleen, regardless of how long we actually had the period of restricted activity. We restrict activity 2 to 3 months at this point.

R. Powell (Mobile, AL): With the data that you have collected on the use of ultrasound scan, have you modified your follow-up in obtaining fewer ultrasound

LYNCH ET AL

scans, especially in the grade III and IV injuries. The biggest problem, at least related to adolescence, is contact sports, mainly football. Do you modify your activity restrictions in these patients?

J.M. Lynch (response): What we are doing now is bringing the patients back at the points defined in the paper Rather than obtaining ultrasound scans every 4 weeks we will obtain it, if it is a grade II lesion, every 8 weeks. We do bring them back at 4 weeks for a physical examination just to be sure there are no symptoms. We are obtaining the ultrasound scan at the times defined in the paper.

In regard to sending them back to physical activity, you may know, if any of you have been to Pittsburgh, everybody in Pittsburgh plays football-at least every boy plays football and hockey, and some girls, too. So it is a hard thing to do to keep them out of sports for a long term. Even in the grade II and III injuries, we try to keep them out for a few more months after we have docu- mented the healing. I have had parents ask me if they could use kidney pads, flak jackets, all sorts of things.

R. Miller (Jackson, MS): Do you worry about splenic function in these higher grade injuries? We have several patients with something seen on ultrasound scan but nothing on the radioisotope scan. Should we keep these patients on penicillin?

J.M. Lynch (response): Studies performed mostly in the adult literature show that you need approximately 30% or one third of your spleen to keep normal immuno- logic function. So we estimate that on the initial CT.

If we don’t feel there is that amount of tissue well vascularized, those patients, we feel, should be immu- nized. Basically, the grade IV splenic injuries should be

immunized and the patients sent home on prophylactic penicillin, although I don’t know what the appropriate length of time to keep them on the penicillin is.

l? Colombani (Baltimore, MD): During this study. how many patients actually underwent surgery for repair or splenectomy? To reiterate what Rick Pearl said, in Baltimore over the last 15 years or so we have not had any patients return to the hospital with complications from their ruptured spleen. As a result, our patients go home on day 5. They have no follow-up studies, and they are allowed to return to full activity around 8 weeks after their injury. We have not had any problems.

J.M, Lynch (response): There were eight patients who underwent surgery in this period of time. The indication for operation of those patients was the inability to stabilize vital signs. They were all performed within 4 hours of admission to the hospital. There were no patients that underwent surgery after that 4-hour period. We transfused approximately 5% of our patients. The average transfusions have been 10 cc/kg, so maybe we don’t even need to transfuse them if we allow their Hematocrits to go a little lower.

B. Saks (Springfield, MA): We notice a number of our children have infectious mononucleosis. And when you have a trauma that seems less than the injury, we always check for infectious mononucleosis just to be sure that is not involved. Do you have any information on how much mononucleosis was seen in your group?

J.M. Lynch (response): I think we have seen two. One was in a football player and one was in a girl. She arrested from spontaneous rupture of her spleen from mononucleo- sis. But I don’t have any information other than those.