From the *Jefferson Medical College of Thomas Jefferson University, Philadelphia, PA; †Department of Ophthalmology, Children’s Hospital Colorado, Aurora, CO; ‡Department of Ophthalmology, Royal Children’s Hospital, Victoria; §Department of Paediatrics, University of Melbourne, Melbourne, Australia; and ∥Pediatric Ophthalmology and Ocular Genetics, Wills Eye Institute, Philadelphia, PA.
Manuscript received December 22, 2011; accepted March 12, 2012.
The study was supported in part by the Foerderer Fund.
The authors report no conflicts of interest.
Reprints: Alex V. Levin, MD, MHSc, Wills Eye Institute, 840 Walnut St, Suite 1210, Philadelphia, PA 19107-5109. E-mail email@example.com.
Abstract: Retinal hemorrhages in children in the absence of risk factors are regarded to be pathognomonic of shaken baby syndrome or other nonaccidental injuries. The physician must decide whether the retinal hemorrhages in children without risk factors are due to abuse or cardiopulmonary resuscitation with chest compression (CPR-CC). The objective of this study was to determine if CPR-CC can lead to retinal hemorrhages in children. Twenty-two patients who received in-hospital CPR-CC between February 15, 1990, and June 15, 1990, were enrolled. Pediatric ophthalmology fellows carried a code beeper and responded to calls for cardiopulmonary arrest situations. At the scene of CPR-CC, an indirect funduscopic examination was conducted for presence of retinal hemorrhages in the posterior pole. Follow-up examinations were performed at 24 and 72 hours. Of the 22 patients, 6 (27%) had retinal hemorrhages at the time of CPR-CC. Of these 6 patients, 5 had risk factors for retinal hemorrhages. The sixth patient had no risk factors and may have represented the only true case of retinal hemorrhages due to CPR-CC. Retinal hemorrhages are uncommon findings after CPR-CC. Retinal hemorrhages that are found after CPR-CC usually occur in the presence of other risk factors for hemorrhage with a mild hemorrhagic retinopathy in the posterior pole.
Cardiopulmonary resuscitation with chest compressions (CPR-CCs) may be a part of a home or hospital attempt to resuscitate a child who is unresponsive. The possibility that CPR-CC can cause retinal hemorrhage has serious legal, social, and medical implications. Retinal hemorrhages are often found in children who are victims of abusive head trauma (AHT), in particular AHT that is characterized by repetitive acceleration-deceleration forces with or without blunt head impact (shaken baby syndrome, SBS) and with approximately two thirds showing numerous and multilayered hemorrhages that extend to the periphery of the retina. These retinal hemorrhages have no specific anatomic pattern, and often involve the entire retina. Retinal hemorrhages are found bilaterally in approximately 85% of SBS victims.1 Macular retinoschisis, a splitting of the retinal layers, is also a common finding in SBS victims.2–4 When a child presents with retinal hemorrhages in the absence of other risk factors (eg, meningitis, hypertension, coagulopathies, endocarditis, sepsis, and vasculitis), the physician must decide whether the retinal hemorrhages are due to abuse or the alleged attempt at CPR-CC. Although prior studies have demonstrated that the severe retinal hemorrhages characteristic of abusive head injury do not occur because of CPR-CC, in almost every study the retinal status before CPR-CC is unknown.5 We examined the retina of children at the time of CPR-CC in an attempt to visualize the retina as close to pre-CPR as possible, and in the short period thereafter, in attempt to better assess whether retinal hemorrhage actually resulted from the CPR-CC.
MATERIALS AND METHODS
Children who received CPR-CC during their stay at The Hospital for Sick Children, Toronto, between February 15, 1990, and June 15, 1990, were enrolled in this study. Pediatric ophthalmology fellows (R.W.E., J.E.E., and A.V.L.) carried a code pager and responded to all hospital-wide calls for cardiopulmonary arrest situations regardless of the time of day. Upon arrival to the scene where CPR-CC was being performed, the ophthalmologist immediately conducted indirect funduscopic examination for presence of retinal hemorrhages and other abnormalities in the posterior pole. Pharmacologic dilation was not used as the pupils in each case were fixed and mid-dilated given the extremis of the patient. CPR-CC was not interrupted for the eye examination if CPR-CC was in progress. Follow-up examinations were conducted at 24 and 72 hours. The study was approved by the Research Ethics Board of The Hospital for Sick Children.
Twenty-two patients were enrolled in the study, representing 90% of the children who received CPR-CC at The Hospital for Sick Children during the study period. Missed cases occurred only due to unavailability of the pediatric ophthalmology fellows. Of the 22 patients, 6 (27%) had retinal hemorrhages. In these 6 patients, retinal hemorrhages at the time of CPR-CC were observed in 1 case of SBS, 1 probable case of child abuse, 1 severe motor vehicle accident head trauma, 1 infant with zone 1 stage 1 retinopathy of prematurity (ROP), and 1 infant with zone 3 ROP. The sixth patient, a 10-month-old baby, was found to have scattered peripapillary hemorrhages without any other existing risk factors for retinal hemorrhage. The baby was found apneic and pulseless by a babysitter with no given history of trauma and received 30 minutes of CPR-CC. Of the 16 patients who had normal retinal findings after CPR-CC, there were 6 cases of congenital heart disease, 2 cases of aspiration, 1 child abuse (not SBS), 1 third-degree heart block, and 6 with unknown history. No patient had new retinal hemorrhages at the subsequent examinations.
There were 6 patients who coincidentally had pre-CPR examinations at The Hospital for Sick Children. In three patients, the initial examination was not performed temporally close enough to the CPR to be valuable as the length of intervening time was greater than several weeks, a period which was clearly sufficient for preexisting retinal hemorrhage to resolve. The baby with stage 3 ROP had hemorrhages after CPR-CC which had been documented before CPR-CC. Two other patients had normal examinations before and after CPR-CC. One of these two patients went on to develop a few macular hemorrhages weeks after CPR-CC. These hemorrhages were due to ongoing thrombocytopenia and heparin coagulopathy, risk factors which were present at the time of CPR-CC. The patient with congenital heart disease had no retinal hemorrhages despite 6 cumulative hours of CPR-CC in the presence of endocarditis, sepsis, thrombocytopenia, and disseminated intravascular coagulation.
Determining the relationship between retinal hemorrhages and CPR-CC has important legal and medical implications. When physicians examine children who have received CPR-CC and present with retinal hemorrhages, the physicians must determine whether the retinal hemorrhages are caused by CPR-CC or by nonaccidental injuries. Children who are victims of AHT may be subjected to CPR-CC when their injuries are severe. If the retinal hemorrhage is attributed to CPR-CC, the diagnosis of abuse may be overlooked or given less weight, thus preventing appropriate intervention and protection of the child from the potential of further harm. The ideal study would include before and after CPR-CC examinations. This is, of course, not practical as one cannot identify in advance which children will need CPR-CC. Our study is unique in that we examined children at the time of CPR-CC, thus getting as close to a pre-CPR examination as practically possible. With this method, we were able to record the appearance of the retina before it could possibly be damaged because of CPR-CC. We then examined the children at intervals thereafter to identify retinal hemorrhage that may develop after CPR-CC.
In our study, 6 of 22 children had retinal hemorrhages noted at the time of CPR-CC. In no case was additional retinal hemorrhage identified on follow-up examinations within 72 hours of CPR-CC. Five of the six patients with retinal hemorrhage were either abused or had known risk factors (fatal motor vehicle accident6 or active ROP7,8) independent of CPR-CC for retinal hemorrhage, and in one case, documentation of the retinal hemorrhages before CPR-CC. Given that the hemorrhages were present at the time of CPR-CC, this suggests that they were preexisting. Only one patient had retinal hemorrhages at the time of CPR-CC in the absence of other risk factors that could represent the only true case of retinal hemorrhages due to CPR-CC, although, once again the presence of hemorrhages at the time of CPR-CC suggests they were preexisting. Complete social and home scene investigation were not done. This case may also represent a missed diagnosis of child abuse. Ten patients with known risk factors did not have retinal hemorrhages during or after CPR-CC.
In cases where retinal hemorrhages were seen after CPR-CC in nonabused children, the type of hemorrhage was distinctly different from the severe retinal hemorrhages seen in approximately two thirds of the SBS cases.1 Our patients all only had a small number of dot and flame hemorrhages concentrated in the peripapillary area or along the major vascular arcades occasionally with a few scattered hemorrhages of the same type in the macula. Small disc hemorrhages also occurred.
Other studies also reached similar conclusions. Odom et al9 performed a study involving 43 patients, only one of whom had retinal hemorrhages after CPR-CC, in the presence of prolonged prothrombin time (22.9 seconds) and partial thromboplastin time (78 seconds) and low platelet count (91,000 platelets/μL). The remaining 42 did not have retinal hemorrhages despite having a coagulopathy and/or thrombocytopenia. Gilliland and Luckenbach10 studied the retina of 169 children at autopsy to examine the connection between retinal hemorrhage and resuscitation. Seventy of these children were found to have retinal hemorrhage. Of these 70 children, 61 had received prolonged resuscitation only one of which was identified as not having a risk factor known to be associated with retinal hemorrhage. That one child was the third sibling to die without apparent cause, and there was history of abuse in the family. As this was a postmortem study, we do not know the clinical pattern of the hemorrhages but the authors conclude “retinal hemorrhages were not found in the absence of injuries or natural disease known to cause hemorrhages in our series.”
Goetting and Sowa published a study suggesting that retinal hemorrhages can result from CPR-CC. They examined 20 patients who allegedly had no other risk factors for retinal hemorrhage.11 Three patients had retinal hemorrhage. The study unfortunately had several flaws. We are not told the risk factors which were excluded, examinations were not conducted by an ophthalmologist, and dilated fundoscopy was not always performed. In addition, one patient was a near drowning. The most common cause of death in drowning is laryngospasm. The struggling victim is left trying to breathe while struggling against a closed glottis. Valsalva retinopathy with retinal hemorrhage is a well-described entity. A second child died 38 hours after admission with possible sepsis, which may also result in retinal hemorrhage. The third child was examined minutes before his third and last cardiac arrest. At that time, he had no retinal hemorrhage. On reexamination after 12 minutes of CPR-CC, he had two small hemorrhages in the peripapillary area of one eye. One paper describes more extensive retinal hemorrhage attributed to CPR-CC.12 These retinal hemorrhages occurred in the setting of multiple independent risk factors.
In summary, our study supports the existing literature in concluding the retinal hemorrhage is uncommon after CPR-CC, and when it occurs, it does so in the setting of coexisting risk factors for hemorrhage with a very mild hemorrhagic retinopathy confined to the posterior pole.
The authors thank Research Assistant, Dr Rizwan Alvi, for the assistance in the preparation of this article. This is dedicated to the memory of Dr J. Donald Morin, who inspired the authors to do this study and provided the leadership to allow for its completion.
1. Morad Y, Kim Y, Armstrong D, et al. Correlation between retinal abnormalities and intracranial abnormalities in the shaken baby syndrome. Am J Ophthalmol
. 2002; 134: 354–359.
2. Levin AV. Retinal hemorrhages: advances in understanding. Pediatr Clin North Am
. 2009; 56 (2): 333–344.
3. Togioka BM, Arnold MA, Bathurst MA, et al. Retinal hemorrhages and shaken baby syndrome: an evidence-based review. J Emerg Med
. 2009: 37 (1): 98–106.
4. Greenwald M, Weiss A, Oesterle C, et al. Traumatic retinoschisis in battered babies. Ophthalmology
. 1986; 93 (5): 618–625.
5. Levin AV. Retinal haemorrhage and child abuse. In: David T, ed. Recent Advances in Paediatrics
. London: Churchill Livingstone; 2000: 151–219.
6. Kivlin JD, Currie ML, Greenbaum VJ, et al. Retinal hemorrhages in children following fatal motor vehicle crashes: a case series. Arch Ophthalmol
. 2008; 126 (6): 800–804.
7. Lim Z, Tehrani NN, Levin AV. Retinal haemorrhages in a preterm infant following screening examination for retinopathy of prematurity. Br J Ophthalmol
. 2006; 90 (6): 799–800.
8. Polito A, Eong K, Repka M, et al. Bilateral retinal hemorrhages in a preterm infant with retinopathy of prematurity immediately following cardiopulmonary resuscitation. Arch Ophthalmol
. 2001; 119: 913–914.
9. Odom A, Christ E, Kerr N, et al. Prevalence of retinal hemorrhages in pediatric patients after in-hospital cardiopulmonary resuscitation: a prospective study. Pediatrics
. 1997; 99 (6): e3.
10. Gilliland MG, Luckenbach MW. Are retinal hemorrhages found after resuscitation attempts? A study of the eyes of 169 children. Am J Forensic Med Pathol
. 1993; 14 (3): 187–192.
11. Goetting M, Sowa B. Retinal haemorrhage after cardiopulmonary resuscitation in children: an etiologic evaluation. Pediatrics
. 1990; 85 (4): 585–588.
12. Weedn VW, Mansour AM, Nichols MM. Retinal hemorrhage in an infant after cardiopulmonary resuscitation. Am J Forensic Med Pathol
. 1990; 11 (1): 79–82.
© 2013 by Lippincott Williams & Wilkins.