Neonatal Stroke and Maternal GBS Infection—Not Always What You Think

Patient X was born on March 22, 2020. By July of 2020, it had become clear that he was not using the left side of his body in a normal fashion. Imaging performed late that month revealed a right hemisphere stroke, likely neonatal/perinatal in origin. Given the absence of significant trauma or asphyxia, this did not fit the usual profile for the birth injury cases we pursue, and indeed was one which many of us may decline at the time of the initial interview. However, patient X’s case is one which should lead us to rethink how we approach cases involving GBS and stroke. At a minimum, we should take a look at what occurred during the labor and delivery beyond the prism of an asphyxia or traumatic event.

We have all probably considered taking an ischemic arterial neonatal/perinatal stroke case and have maybe even found expert support for trauma or asphyxia causing same. In those cases, it is almost guaranteed that defense experts will be found who will state that asphyxia cannot cause a unilateral injury, and that any trauma would have to be quite severe to even potentially result in same. For patient X, neither of those potential etiologies was in play, as the labor and delivery was without any unusual trauma, and the baby at birth, and indeed for a few months beyond, was well appearing and normal. There was no hint, in fact, of what was to come.

The Culprit – Untreated GBS

An interesting feature of this L&D was the fact that mom came in with unknown GBS status, having ruptured membranes at about 34 weeks. GBS testing is usually done around 35 weeks of gestation, and so for those women with PROM and labor prior to that time, GBS status will often be unknown. In this case, during the labor she became progressively more clinically ill, eventually developing a fever and fetal tachycardia in response to same. Initially puzzling was the fact that this occurred despite the administration of antibiotics from the moment she arrived at L&D.

After delivery, mom’s placenta and cord were sent to pathology and found to be markedly inflamed, with the offending organism identified as GBS. These findings would normally play well into what is often the defense position in stroke cases, namely that inflammation is a known precursor to perinatal/neonatal stroke. Indeed, in this case, it was the position of plaintiff’s experts that inflammation was the cause of patient X’s stroke, with the severe inflammation being pro-thrombic, and there being a direct pathway for embolism of a thrombus from the umbilical cord to the neonatal brain. So, one might ask, why is any of that good for this plaintiff, and why would anyone take such a case? The answer lies in the near-total lack of understanding on the part of the defendant OB as to one very basic principle of GBS prophylaxis. So basic, in fact, that it was alarming to learn just how many practitioners were not aware of it until very recently.  

“If the prenatal GBS culture is unknown when labor starts, intrapartum antibiotic prophylaxis is indicated for women who have risk factors for GBS EOD. At-risk women include those who present in labor with a substantial risk of preterm birth, who have preterm pre-labor rupture of membranes (PPROM)[.]… Intravenous penicillin remains the agent of choice for intrapartum prophylaxis, with intravenous ampicillin as an acceptable alternative.”

That guideline has been accepted for many years but fails to address one very important scenario – the penicillin allergic patient. Patients who require prophylaxis but who have a serious allergy to penicillin require alternative medications to help prevent GBS infection in their baby. The question in this case, as it should be in all cases, is which one:

Q: And it’s your understanding, based upon your recent review, that if the patient is of unknown GBS status and is strongly allergic to penicillin, then it is indicated that one use Clindamycin?

A: Okay, if the patient has a penicillin allergy, the literature says you might think about cephalosporin; however, in my training, if you’ve had a penicillin allergy, I’ve always gone right to Clindamycin.

(Deposition testimony of Defendant obstetrician.)

In fact, the views expressed above have not been valid since about 2010, when it was becoming increasingly appreciated that GBS organisms were developing a resistance to clindamycin. Indeed, as of the time of this delivery in the early 2020s, it had been well established that up to 40% of GBS bacteria were resistant to clindamycin, leading to a change in the guidelines years before the birth in question in this case. For those women whose GBS status was unknown (as here) or known to be positive and their organism resistant to Clindamycin, the standard of practice was to use Vancomycin as the only proven effective drug for prophylaxis for this disease. CDC,Antimicrobial Resistance: 2019 Antibiotic Resistance Threats Report, U.S. Dep’t of Health and Human Services (2019); American Academy of Pediatrics & The American College of Obstetricians and Gynecologists Guidelines for Perinatal Care, 419-420 (7^th ed. 2012) / ); American Academy of Pediatrics & The American College of Obstetricians and Gynecologists Guidelines for Perinatal Care, (8^th ed. 2017).

This simple rule was unknown to the defendant in this case, leading him to give an essentially worthless antibiotic to the mom, and allowing her GBS infection to progress unchecked during her nearly 30 hours in labor. It was no surprise to find that she became clinically ill during that timeframe, and similarly almost expected that her cord and placenta would be severely infected and inflamed from unchecked GBS. Pathology examination of the placenta and cord not only identified the infection but also determined it to be Clindamycin-resistant GBS.

It bears mentioning that in this particular case, but for the negligence of the nursing staff at the hospital, the physician negligence under discussion here would never have occurred. There had been an order for GBS testing given during a hospitalization some 5 days earlier. For reasons that never became clear, it was never followed through on by the nursing staff. This failure was of great import since, in view of the resistance issue, it was well established at laboratories throughout the country that any GBS sample sent for a patient who has a serious penicillin allergy must also undergo sensitivity testing. Providers/nurses are to indicate the allergy on the lab requisition form (in some record systems, the notation of allergy will be reviewed by pathology directly), which will trigger an automatic sensitivity being performed. Laura Filkins et al., Guidelines for the Detection and Identification of Group B Streptococcus, Am. Soc. For Microbiology (March 10, 2020, updated July 23, 2021).

Presumably, even for this defendant obstetrician, notice of resistance to clindamycin would have led to alternative drug use for prophylaxis (he testified as such in fact). The failure to properly provide antibiotic prophylaxis to this patient led to clinical illness related to her GBS colonization and concurrent worsening infection/inflammation of her placenta and cord.  

When Did the Stroke Occur?

Timing a stroke as having occurred around the time of birth involves both clinical and imaging observations. From the clinical perspective, an interesting feature of perinatal stroke that we should all keep in mind is that the large majority of babies thought to have sustained an “idiopathic” stroke are full term. This baby, as we know, was born at 35 weeks, leading one to look for a specific cause for this event. Here, the only event of note was the severe inflammation, leading plaintiff’s stroke neurologist to testify that there was no other reasonable explanation for this stroke. 

From an imaging standpoint, MRI, if done during the neonatal period, may be extremely valuable as a tool to determine timing. Here, however, as referenced above, there was no MRI done during the neonatal period. There was, however, an ultrasound done at 6 days of life, which was read as normal by the hospital radiologist. Were this true, it would be somewhat problematic since evidence of a stroke during delivery can often be seen even on U/S by that time. Like many cases, this one highlighted the importance of having your own expert read the films. When the expert reviewer, a very well-known pediatric neuro-radiologist, looked at the U/S he identified clear evidence of edema consistent with stroke around the time of delivery.

Why “Normal” Until 4 Months of Age?

If you ever have a potential perinatal/neonatal stroke case, do not be put off by the fact that the child did well at birth and didn’t show signs of stroke until later in life. In this case, those abnormal findings were present by about 4 months of age, and initially, this confounded those of us ignorant of normal infant neurology. Such a course is, in fact, expected, and is explained by an expert in stroke neurology below:

Q:… And then it says: “The parents only started noticing this at four months of age. Is that right?

A: Yes.

Q: Would you have expected other symptoms to reveal themselves before four months of age?

A: No, this is actually a very classic presentation.

Q: Okay. Why does it take so long for the symptoms to present themselves if a child has a perinatal arterial stroke?

A: That’s a good question. So when babies are born, they aren’t using the portions of their brain that are supratentorial, so they do not have purposeful hand movements or leg movements. They are moving them, but they are not with purpose. And as the brain matures, you start to use your extremities purposefully and that – the combination of that development is – – those skills develop based on how your brain in the motor cortex works and if your motor cortex is injured, you won’t start noticing the issue until you start needing to use the motor cortex. So right around 4 or 5 months is, as you may be aware, a baby starts reaching for toys with purpose as opposed to swatting themselves, so, you know, around that time, they are using their brains to say I see something and I want to grab it and it’s with purpose…

A discussion of the concept described in the above testimony can be found in many pediatric neurology publications, including the below. (If you review it, don’t let one of the authors put you off to the concepts discussed therein). See Janette Mailo et al., What Do We Know About Perinatal Stroke? A Review of Current Practices, Outcomes, and Future Decisions Pediatric Stroke (2021).

We all have in our minds the prototypical models for a winnable birth injury case. Bad tapes, delay in c-section, a cord pH of 6.8 and BE of -20 is one obvious frequent scenario. However, if one does enough of these cases, you come to realize that that model is not always going to be there for you, but that something else, maybe a bit less obvious and even, as here, counterintuitive, might surprise you in where they lead, and what you can do for your client. 

This article originally appeared in the Summer 2025 edition of the American Association for Justice’s BTLG Journal.