Slides
Transcript
Alright, thank you and good morning. Just to start out, as stated, I am a neuroradiologist, so I approached this topic with the expertise of someone who diagnoses and treats CSF leaks. I am not an OB; I am not an anesthesiologist. Anytime you’re talking about the care of a pregnant patient planning for labor and delivery, there needs to be a multi-disciplinary conference and consideration.
Okay, so an overview of the topics we’ll cover today. I’ll talk about the first presentation of spontaneous leaks during pregnancy and the options for diagnosis and management, options for labor and delivery anesthesia, and a few examples of some postpartum leak cases. And then lastly, some pharmacotherapies and considerations that we need to keep in mind during the pregnancy and lactation.
So, anybody that has experienced pregnancy themselves or has been in close proximity and witnessed this knows that there are significant physiologic changes. Of course, we all see the big pregnant belly. I think it’s always very fascinating to see what this looks like on the inside – all these organs that are normally in your abdomen and pelvis are smashed up here, smashed down here.
What’s probably apparent to the physicians in the room, but I want to point out, is that this is a CT scan, this not an MRI. My understanding is that this image was obtained in the setting of a rather severe trauma in a 37-week pregnancy. However, it was deemed that obtaining this study was in the best interest of the mother and the fetus to rule out severe injury. So, irradiating scans are not a complete contraindication but have to be taken into consideration, and that’ll come up a little later.
This was from a fascinating study that just came out in Nature Neuroscience in which a remarkably patient and not claustrophobic woman was scanned 26 times over the course of preconception, through her pregnancy, and after. This person has self-identified. She herself is a neuroscientist, and this is an N of one. But what we saw are these amazing changes that, of course, happen in the body but also in the central nervous system.
So, in particular, the CSF – this was, again, brain MRI – so we’re looking at changes in ventricular volume: a 60% increase in CSF volume over the course of a pregnancy. As we’ve discussed, Monroe-Kellie doctrine – there’s only so much space in the skull, so this is accompanied by changes in overall brain volume, which gets smaller, but the white matter tracts get much denser, almost like the brain becomes lean and mean.
Despite this huge increase in CSF volume, the opening pressures in women, even through their third trimester, remain normal. So, there is some kind of accommodation that takes place, that even though we have this greatly increased volume, there is not an increase in pressure. However, we also have this marked increase in sex steroids.
This is, again, data from that first study I just showed about, you know, there are very small fluctuations over the normal menstrual cycle, but during pregnancy, huge increases.
We know that this state that happens in pregnancy and a non-pregnant person can be associated with elevated intracranial pressure, either in the setting of an increased body mass index, increased sex hormones. So, is pregnancy a possible risk? Could we see this popping-the-balloon phenomenon that we sometimes see in these patients with obesity or with chronically elevated intracranial pressures?
However, there have been numerous case reports of patients with spontaneous improvement during pregnancy, spontaneous improvement after delivery, or possible recurrence in subsequent pregnancies. This paper by Rohatgi et al. – some of the senior authors include Dr. Matthew Robbins, Dr. Levi Chazen, which are also leaders in our field – and this is the most comprehensive review and case collection. As in most things, we can’t do studies in pregnant women. We just have to study case collections.
Something that they brought up is that there is this extensive remodeling of the epidural space that happens in pregnancy. This was a fascinating study. I didn’t even know that this was something that could be done. This is epiduroscopy. So, as they place the epidural needle, you can actually place a small scope or camera in the epidural space, and they were able to quantify these dramatic changes. This is over the course of pregnancy, sort of mid-pregnancy and at the end of pregnancy. There is proliferation of epidural fat that we probably could have predicted, but also this extensive change in the epidural vasculature.
So possibly, as in that case that Dr. Schievink presented earlier, we see that there is this significant remodeling. Could this spontaneously improve leaks? I can’t endorse pregnancy as a treatment for leaks, but we know that there is significant remodeling of the epidural space. In this study as well, they report that one of their patients felt better when she was pregnant. It was thought to possibly be related to this process, or just even the gravid uterus forcing more CSF back towards the intracranial compartment.
Okay, so what do we do when a patient presents with new suspected SIH during pregnancy? So, our workhorse studies here are MRIs, as they always are. The difference is that we can’t give the gadolinium contrast, so we can still do non-contrast MRI of the brain. We can still look at our, you know, Bern criteria here of our sellar distances. We can actually tell some venous engorgement also on non-contrast imaging pretty well, and then our heavily T2-weighted fat-saturated image that we normally use as part of our CSF leak protocol. This is all because gadolinium is renally excreted, which in a person is not a problem. During a pregnancy, it is a problem because the fetus will also excrete the gadolinium into the amniotic fluid and continue to swallow that, so it’s not eliminated from the pregnant patient.
These are some images from a case that we had here. So, this was a woman presenting with new severe postural headache in pregnancy. She had a Bern score of four. We can see some narrowing of these key distances, slumping of the brain stem against the clivus, descent of the cerebellar tonsils. She had a few meningeal diverticula, no extensive SLEC, but maybe a little bit of that epidural fluid signal – very subtle dinosaur tail sign. So, we found this, we’re suspicious. What can we do for these patients?
The first line is going to be conservative measures. You can try bed rest, hydration, steroids. When it comes to pharmacologic treatments, we’re a little bit limited, right, because you can’t give high-dose caffeine to a pregnant woman. The obstetric societies recommend less than 200 milligrams a day. I think that’s about two grande Starbucks, so not too little but not much. And then for NSAIDs – the non-steroidal anti-inflammatories, things like ibuprofen – these can be used sparingly in early pregnancy. They are not recommended to be used after 20 weeks. The reason why is, again, you don’t want to affect that fetal renal function that could influence amniotic fluid volumes.
Second line, you can do invasive treatments. So, non-image-guided empiric blood patches are going to be the workhorse here, usually administered by anesthesia. Another suggested modality has been greater occipital nerve blocks – so not targeting the leak itself but seeking to make some of the associated pain tolerable. And then in severe cases, you’ll consider the risks of studies that use radiation, either DSM or CTM, or image-guided epidural patching.
You know, these cases are rare in general, but if you’re talking about something that’s subdural hemorrhages, dural venous thrombosis, something that’s going to threaten the life of the mother or the viability of the pregnancy, maybe it’s indicated. And then the radiation risk is lower later on in pregnancy once organogenesis is completed, about week 20 – thought to be much lower risk to the fetus.
Okay, this is the conclusion of that patient’s story that I showed you earlier. So, she did receive an empiric blood patch that was able to get her through the pregnancy. However, after she delivered, she had persistent symptoms. At that time, she was taken for a myelogram, and this very small, sort of slow extravasation of contrast was seen – thought to be maybe a small lateral tear. She received targeted epidural fibrin patching with Dr. Callen, and now, over a year out, she continues to do very well.
Considerations for labor and delivery anesthesia. So, this is hard because, again, we don’t have good studies, but it is thought that for patients prior, if they have a leak going into a pregnancy or develop a new leak during pregnancy, they should not be at a higher inherent risk for an iatrogenic leak related to spinal or epidural anesthesia. However, we do need to avoid areas of possible pre-existing dural injury. And then, of course, this caveat is not for those with connective tissue disease who might be more prone to a dural injury in general.
This is a schematic of the different needles that we can use. I think a couple of people have shown a version of this. Basically, yes, we want to keep with those pencil-point, non-cutting needles for spinal anesthesia, making the dural puncture, and these needles used for epidural anesthesia are quite large – you have to pass the catheter through there.
These are just some different options. What I wanted to point out – we’re familiar with the standard epidural spinal anesthesia, whether in the setting of a cesarean section, which is when it’s commonly used, or in combination with a spinal epidural – something called the combined spinal epidural, or CSE. This is the modality of choice for some OB anesthetists. It’s certainly used here with a very high frequency. At first, this was counterintuitive to me. Why would you want to purposely puncture the dura? But the idea is that you can make that spinal anesthesia puncture with a very small, atraumatic needle in a planned and controlled way and avoid an injury from the larger needle.
Of course, general anesthesia remains an option but is less optimal. The mother will not be conscious for delivery, and it actually does not provide good post-operative pain control.
This is a schematic of the combined spinal anesthesia. These are usually done through a needle-through-needle technique. There are a few different designs: one where the spinal needle comes out from this common tip or from this back wall tip. The idea here is that this spinal needle can actually be used as a depth check, that you can get CSF return and know not to further advance that epidural needle.
There’s a little bit of controversy in this area in the anesthesia literature, but one report is that there is a 2.5 greater risk of post dural puncture headache and dural injury when you’re doing an epidural alone versus the combined spinal anesthesia.
Quickly showing some cases on the two extremes of what can happen related to an iatrogenic leak during labor and delivery anesthesia. This was a 35-year-old who had an epidural placed in the setting of a labor induction. It was placed at L4 – L5. She subsequently had a precipitous labor and an unattended delivery. She had a worsening headache when upright. She underwent bedside patching with anesthesia twice. She was discharged home still with a severe postural headache and was unable to care for her infant.
So, she came back to the emergency department. At this point, we asked for brain and spine imaging. She could now get contrast because she’s postpartum. It shows very classic findings of severe brain sag. I believe her Bern was six or seven. I have purposely not shown the lumbar spine, which was a bit of a mess after two epidural blood patches there. What I wanted to point out here is that she did have quite a significant leak with an epidural fluid collection extending into the upper thoracic spine and cervicothoracic junction.
We then took her for targeted fibrin epidural patching right at the site at L4 – L5. You can appreciate how this would be a difficult procedure to do without imaging guidance. I believe this required a 15 cm needle. Within a few months after that, she was completely resolved. Brain findings are now normal, and that spinal collection is resolved. She is doing well, currently in nursing school.
Another case – this has been previously published by Doctors Callen, Lennarson, and Carroll. This woman received a labor epidural when she was 24 years old. She had an orthostatic headache at that time. She improved after bed rest, and went on to live her life headache-free for 12 years. She then developed a new persistent daily headache for six years – so we’re talking 18 years after her labor epidural, and her MRI brain was negative. The only finding on her spine MRI was this teeny tiny little bleb.
When Dr. Callen took her for a procedure – this was a targeted patch – we saw this phenomenon called intravasation. We usually think of extravasation, where the fluid comes out. For a number of leaks, for whatever reason, we can only see the contrast leaking in. She received profound but temporary relief from this procedure. Dr. Lennarson then took her to the operating room, and corresponding to that little defect, saw this blood pouching out. That was repaired, and she is subsequently doing very well.
As discussed for the caffeine and NSAIDS, we also have to look at medications for rebound intracranial hypertension and if those are safe in pregnancy and breastfeeding. The good news is that they are. The only one that really carries additional risk is topiramate. I’ll point out this drug and lactation database that is available through the National Institutes of Health. This is open and accessible to anyone, so if you have questions about whether a drug is safe in pregnancy, in lactation, and at what doses, this is a great source of information.
This is another busy slide, but again, just wanted to point out that most of these agents are safe. You have to do some additional monitoring of potassium levels and, as well, be careful to avoid dehydration as that could have potential effects on amniotic fluid levels or breast milk production and lactation.
In conclusion, pregnancy, labor, and delivery confer distinct risks to develop or exacerbate a spinal CSF leak. Diagnostic and treatment options are available when pregnant or lactating. Make an informed decision regarding risks and benefits of OB anesthesia and look for a supportive team. Thank you very much.