Sunday, December 29, 2019

Shock, bradycardia, ST Elevation in V1 and V2. Activate the Cath Lab?

A 60-something with h/o cirrhosis and diabetes called 911 because he felt sick and was unable to move his lower extremities

On arrival he was bradycardic and hypotensive.

He stated that starting approximately 7 hours prior the he felt that he was unable to feel his extremities.  At some point after that he contacted his neighbor who came to check on him and called 911.  On arrival to the stabilization room he says he can feel his extremities and and states that he just generally feels unwell.  He denies any chest pain or shortness of breath.


EMS reports that when they arrived to the scene he had a heart rate in the 40s with very weak and thready peripheral pulses and altered mental status.  He had a syncopal episode while getting into the EMS cart.  He maintained pulses but was ashen and diaphoretic.  He was largely unresponsive to voice during his transport to the hospital.

BP systolic 60.

Ultrasound revealed global dilation of all 4 chambers of the heart with poor contractility, a IVC that measured 2.4 cm, no significant B-lines in lung fields nor large pleural effusions.

An ECG was recorded:
No definite P-waves, Bradycardic, Regular, lots of artifact.
There is ST Elevation in V1 and V2
What do you think?









Comment:  There is also a very wide QRS.  The regular rate suggests: Junctional with aberrancy, or ventricular escape, or sinus bradycardia with an invisible P-wave (this happens in hyperK).  The regular rate rules out atrial fib, unless there is atrial fib with complete AV block and an escape.

Whenever there is bradycardia or a wide QRS, especially BOTH, it is hyperkalemia until proven otherwise.  Hyperkalemia often results in PseudoSTEMI patterns, especially in V1 and V2, but also in inferior leads.

See these other cases of PseudoSTEMI due to hyperkalemia.  Some were unrecognized, resulted in inappropriate cath lab activation, and then resulted in death.  

Based on the initial ECG, the providers activated the Cath Lab.

However very shortly after activation, they received the initial labs.

Unmeasurable lactate over 12
Metabolic acidosis with a pH of 7.08
Bicarb of 11, with no significant respiratory compensation with a PCO2 of 40
Acute renal failure with a potassium of 8.8 and a creatinine of 2.38.

With this information, they deactivated the Cath Lab and focused our treatment on acute hyperkalemia.  

They administered 6 grams of calcium gluconate (equivalent to 2 g of CaCl), 10 units of IV regular insulin, D50, and 10 mg of nebulized albuterol.

A repeat ECG was recorded:
Dramatic improvement
Peaked T-waves remain in all leads, and QRS is still wide, but much better.











Tuesday, December 24, 2019

Christmas Eve Special Gift!! Prehospital Cath Lab Activation: What do you think?

A 60-something male was sent from Dialysis for several days of SOB.  He did not know his medical history well.

The patient was in no distress.

The paramedics recorded an ECG:


Here I magnify the limb leads and precordial leads:

Now of course you see the convex ST elevation in V3 and V4, and the STE in V2 that is preceded by a spike. 


Is this STEMI?

Here is the Computer Interpretation:


















The medics activated the cath lab.  Do you agree? 

I was in the ED, and whenever I hear that there has been a prehospital cath lab activation, I like to go take a look because there are many false positive activations, especially in patients without chest pain.

First: remember that a heart rate that fast is unusual in acute type 1 MI unless there is cardiogenic shock or impending shock.  But the patient appears well.

Second: Look carefully at the limb leads: there are definite flutter waves.

You can also see them easily in lead V1 and V2: see the regular spikes at a rate close to 300?

Third: Whenever you see abnormal ST-T, look at the QRS.  Here you see QS-waves, highly suggestive of old MI with persistent ST Elevation (LV aneurysm morphology -- LVA).  Other reasons for QS-waves are LVH and cardiolmyopathy, and both can have baseline ST Elevation.

----Such baseline STE is often exaggerated by tachycardia.

----Flutter waves can also superimpose and exaggerate the ST Elevation.

My interpretation at the time I saw this: probably a false positive.  

Plan? Let's get an ED ECG and look at old ECGs before confirming cath lab activation.

Here is the ED ECG:
Same


Here is the Previous ECG recorded 2 months prior:
This confirms previous QS-waves with baseline ST Elevation

Here is an echo less than one year prior:
Mildly enlarged left ventricle with severely reduced systolic function.
The estimated ejection fraction is 16%.
Global hypokinesis with minor regional variations
Left ventricular hypertrophy concentric mild.



So it seems that the baseline ST Elevation may be due to LVH instead of LVA, but what matters is that there is baseline STE.

Here is another recent ECG:
It is identical to the presentation ECG!
Atrial Flutter with rapid ventricular response, QS-waves, and exaggerated ST elevation

Cath lab activation was cancelled.

The heart rate was controlled with beta blockers.



===================================
MY Comment by KEN GRAUER, MD (12/24/2019):
===================================
For Dr. Smith’s “Christmas Eve Special Gift” — I’ll limit my comments to brief review of my approach to this rhythm.
  • For clarity in Figure-1 — I’ve lightened and simplified the initial tracing in this case ( = ECG #1, obtained by the paramedics on the scene).

QUESTION: How to go about determining the rhythm?
  • HINT: I reviewed a nearly identical problem-solving approach in MComment in the November 12, 2019 Post on Dr. Smith’s ECG Blog.

Figure-1: The initial ECG shown in this case — obtained by the paramedics in the field. They were initially concerned about an acute STEMI ... (See text).



ANSWER: As I describe in detail in my November 12, 2019 Post — As soon as you establish that this patient is hemodynamically stable — Proceed to Systematic Rhythm Analysis:
  • The rhythm is regular at a rate of ~150/minute (ie, the R-R interval is ~2 large boxes on ECG grid paper — and 300/2 = 150/minute).
  • The rhythm is supraventricular (ie, the QRS is narrow = not more than half a large box in duration).
  • I do not clearly see sinus P waves. It almost looks like there is an upright P wave in lead II — but IF this upright deflection before each QRS in this lead was a P wave, then the PR interval would be fairly short. BOTTOM LINE: We are not certain there are sinus P waves in this tracing.
  • EXTRA CREDIT: Look for sign of atrial activity in the other 11 leads. Look first at lead V1 — since next to lead II, lead V1 is often the best lead to search next for sign of atrial activity. Pull out your trusty pair of CALIPERSDon’t YOU see 2 pointy spikes that regularly fall within the R-R interval of each beat in lead V1? Set your calipers to precisely the interval between 2 of these pointed peaks in lead V1 — and NOTE how you can precisely walk out these spikes at a rate of 300/minute throughout this lead V1. The only rhythm that does this (ie, give you regular atrial activity at ~300/minute) — is AFlutter — and you have just made a definitive diagnosis of the rhythm in Figure-1!
  • SHORTCUT #1: Even before you got to the “Extra Credit” Step — Once you realized that the rhythm was a regular SVT at ~150/minute without clear sign of sinus P waves — You could have said to yourself, Think AFlutter until proven otherwise — and you would have then made a definitive diagnosis of this rhythm in less than 1 minute! (That's because AFlutter is by far the most commonly overlooked cardiac arrhythmia — and the BEST way to never again overlook AFlutter, is to always LOOK FOR IT whenever you have a regular SVT at ~150/minute without clear sign of sinus P waves!).
  • SHORTCUT #2: Also before you got to the "Extra Credit" Step — Remember that when looking for AFlutter (because you have a regular SVT without clear sinus P waves at a rate close to ~150/minute) — it sometimes helps to STEP BACK a little from the tracing. Doing so should highlight the distinct "sawtooth" pattern of AFlutter that is strikingly prominent in the inferior leads of Figure-1. (Learning to recognize this "sawtooth" pattern facilitates recognizing many cases of AFlutter within seconds!).

HAPPY HOLIDAYS to ALL!
  • IF in doubt about what to gift your special friend/colleague who loves ECG interpretation as much as you do — Gift them with a pair of CALIPERS! (Tell them Dr. Ken suggested this! — :)


Monday, December 23, 2019

Wide-complex tachycardia that didn’t follow the rules

Case contributed by Brooks Walsh, an emergency physician and ECG aficionado in Connecticut  (and a few comments by Smith)

A middle-aged patient came to the ED complaining of palpitations, dyspnea, sweating, and chest pressure that radiated to the arms. They had a history of non-ischemic cardiomyopathy (EF 30%), as well as PCI with one stent. Home medications included metoprolol, but no calcium- or sodium-channel blocking agents.

Initial ECG in the ED:

Presenting ECG: Wide-complex tachycardia at a rate about 200.

This is overwhelmingly likely to be ventricular tachycardia, even if only age and medical history are considered. If formal criteria for differentiating VT from SVT are applied, then VT appears even more likely. For example:
Lastly, the total QRS duration in V1, V2 is very long at 200 ms, longer than typical LBBB and again suggests VT.

An old ECG was checked:

Old baseline: Sinus rhythm with a LBBB
The RS duration is 100 ms, and the QRS width is 200 ms,
both markedly long for a LBBB

On one hand, the QRS in aVR in the baseline ECG is markedly different from that in the tachycardic ECG, again supporting a presumption of VT. 

Nevertheless, the widths of both the QRS complex and the RS duration are similar in both the old ECG and the tachycardia. This increased the likelihood that the wide-complex tachycardia was an SVT with a fixed LBBB.
Given some doubts about the ventricular origin of the WCT, the emergency physician elected to try adenosine first. A dose of 12 mg was properly administered, and no changes were noted on the monitor (at the time). The patient was then electrically cardioverted.


Post-conversion: Back to sinus with classic LBBB pattern


The WCT was presumed to be VT, and the patient admitted.

A surprise clue!

Although the emergency physician (along with an APRN and an RN) had not seen any alterations in the rhythm after adenosine was pushed, a paper recording of the episode was subsequently unearthed.
 


The WCT is interrupted by a series of variable-morphology QRS complexes, 
with atrial flutter waves note in II, III, and aVF.  


Detail of Flutter waves
The rate of the flutter waves matches the rate of the WCT (about 200/m), proving that the presenting WCT had been 1:1 atrial flutter. 

The subsequent EP study could not induce VT, only atrial fibrillation. Ablation to prevent recurrent flutter was performed.

This shows that, even if adenosine does not convert the rhythm, it can be useful in diagnosing the rhythm.

Spontaneous 1:1 atrial flutter

Atrial flutter usually has a rate of 300 F/m. The AV node, however, cannot conduct that rate, and typically will only permit 2:1 conduction, producing a usual ventricular rate of 150 bpm.

Atrial flutter conducting at 1:1 can be seen in certain contexts. For example, in a patient taking flecainide for atrial fibrillation, but not a beta-blocker.
Spontaneous atrial flutter with 1:1 conduction (i.e. in the absence of antiarrhythmic use) is rare, with only a handful of case reports in the literature: TanBurkhart, Murthy are 3 relatively recent open-access reports, while Nathwani is a abstract describing an unclear number of subjects with spontaneous 1:1 flutter with those with antiarrythmic-associated 1:1 flutter.

It isn't entirely clear why our patient developed 1:1 conduction with atrial flutter. Certain medical conditions (hyperthyroidism, accessory bypass tract, cathecholamine surge) can provoke it. Exercise can convert atrial flutter from 2:1 conduction to 1:1, apparently due to a combination of accelarated AV conduction and slowed flutter conduction. 

In our patient, the flutter rate of 200 is fairly slow, facilitating 1:1 conduction through the AV node. A slow flutter rate is not invariably associated with spontaneous 1:1 atrial flutter, however. Murthy, for example, described a healthy 50-year-old man who presented with a spontaneous 1:1 flutter at a rate of 280 bpm. 

In addition, the only study that compared spontaneous 1:1 atrial flutter with arrhythmic-associated 1:1 flutter found that the flutter was faster in spontaneous cases.

It makes sense the the atrial flutter rate would be faster in spontaneous cases, as there is not medication slowing that atrial rate, but it seems very unusual that the AV node could conduct at such a fast rate without some other condition (for instance hyperthyroidism).
Nathwani

Conclusion

It is well worth remembering that this was a rare case! 

In most middle-aged patients with a history of cardiomyopathy, a WCT will usually be VT. Furthermore, while specific criteria (e.g. precordial RS duration, aVR morphology) may not be perfectly sensitive, the are more likely to be helpful than suspecting 1:1 atrial flutter in patients not taking sodium-channel-blocking antiarrhythmics.

Comparison of the old and WCT ECGs might have suggested an SVT or atrial flutter at the time of presentation. However, it would not have been prudent for most clinicians to try e.g. a calcium-channel blocker as the first agent. When not absolutely sure, treat as VT!


Adenosine is safe in VT and may be useful in making the diagnosis.




===================================
MY Comment by KEN GRAUER, MD (12/23/2019):
===================================
It is ALWAYS great to welcome the contributions to Dr. Smith’s ECG Blog from Dr. Brooks Walsh — a highly skilled clinician + good friend and colleague who always stimulates conversation on important emergency medicine topics. Today’s case by Brooks is no exception.
  • I agree with many of the SUPERB clinical points emphasized by Brooks.
  • That said — I have a different opinion on several aspects of this case.
  • Definitive answers may be elusive — but my hope is to enhance Brooks’ presentation with important Learning Points for all.
  • BOTTOM LINE  I do not think ( = my opinion) that this wide-complex rhythm failed to “follow the rules”. On the contrary — I thought it followed the rules very nicely!
  • For clarity, in Figure-1 — I have put together and labeled ECG #1 ( = the initial ECG in this case, that was done in the ED) — with ECG #2 (which was the prior ECG found on this patient).

Figure-1: The first 2 ECGs shown in this case (See text). Although unfortunately there is no long lead II rhythm strip — RED arrows in ECG #2 highlight that sinus P waves (the rhythm is sinus bradycardia and arrhythmia) continue throughout the tracing, allowing us to establish with certainty the 7 sinus-conducted beats vs the 2 PVCs in this tracing. My comments below regarding QRS morphology pertain to the sinus-conducted beats in various leads.



I agree entirely with Brooks that ECG #1 shows a regular WCT ( = Wide-Complex Tachycardia) rhythmwithout clear evidence of atrial activity. Although I agree with Brooks that the rate is “about 200/minute” — I feel it important to be more precise with our rate estimation. This is EASY to do — by use of the Every-Other-Beat Method (CLICK HERE and listen to the next few minutes of my ECG Video for explanation of this easy-to-apply Method).
  • By the Every-Other-Beat Method — we can see in lead aVF of ECG #1 that it takes just over 3 large boxes on ECG grid paper (BLUE numbers) — to record 2 beats of the tachycardia (RED numbers). Therefore — HALF the rate, is just under 100 beats minute X 2 = ~190 beats/minute for the rate of the WCT in ECG #1.
I also agree entirely with Brooks that the overwhelming odds for this WCT rhythm without clear evidence of atrial activity favor VT as the diagnosis until proven otherwise! Statistical odds (given that this patient is at least middle-aged, with underlying structural heart disease in the form of a severe cardiomyopathy) that this rhythm is VT are >90% even before you look at the ECG! That said — disagree with the ECG findings selected as “supporting” evidence.
  • Having reviewed for over 3+ decades now all the predictive ECG criteria I have ever been able to “get my hands on” for distinguishing VT from Aberrant Conduction — I’ve developed “My Own Take” on this subject. This reflects my experience and my opinions — which I fully realize others may disagree with! (CLICK HERE for “My Take” on distinguishing VT vs SVT in the diagnosis of Regular WCT rhythms).
  • PEARL #1  Consider depending less on ECG criteria that are based on strict measurements — be this of QRS duration, R-to-S nadir, or other interval durations. This is because the boundaries of QRS onset, S wave nadir (lowest point of the S wave), and distinction between end of the QRS vs beginning of the ST-T wave are not always clear. It is one thing for expert cardiologists in a quiet office to retrospectively review a series of study tracings looking for incriminating measurements — and, another thing to be at the bedside of a crashing patient in a life-threatening tachyarrhythmia in which the speed of all around you is simply not conducive to decision-making based on precise measurements. In my experience — I’ve never been able to precisely measure R-to-S nadir, or QRS segments accurately enough to help during an emergency situation while at the bedside of a “crashing” patient.
  • PEARL #2  It is true that some delay of the initial part of the QRS complex (which I base on appearance rather than strict measurement) is more likely to be seen with VT than with supraventricular rhythms. But this information is best used (in my opinion) as a relative criterion — rather than an exact one. For example — the initial downslope of the S wave in the anterior leads of ECG #1 clearly manifests some delay (S wave downslope in leads V1-thru-V3 in ECG #1 is not a steep vertical line). Seeing this — I thought the slight delay I saw in R-to-S nadir could be consistent with VT — but most definitely was not diagnostic because of the CAVEAT I describe in the next bullet!
  • CAVEAT  All bets are off regarding the finding of initial QRS delay and marked QRS widening IF the patient has severe underlying structural heart disease — as baseline QRS morphology may be quite atypical when the patient has significant heart disease. For example — Note excessive fragmentation (notchingof the QRS complex in multiple leads in ECG #2 (especially in leads II, III, aVL, aVF, and V5). Fragmentation suggests scarring (ie, from prior MI and/or cardiomyopathy). It can therefore be seen from this patient’s prior baseline tracing ( = ECG #2) — that he had been in sinus rhythm with complete LBBB. Because of his severe underlying cardiomyopathy — his baseline QRS complex was extremely wide (attaining ~0.20 second for the sinus-conducted beats in lead V3 of ECG #2). In addition, there was marked fragmentation in ECG #2 — and that could contribute to a delay in R-to-S nadir times.
  • PEARL #3  IF you can be certain of the onset and offset of the QRS complex in lead aVR during a regular WCT rhythm — the finding of an entirely positive monophasic R wave in lead aVR is virtually diagnostic of VT. This is because, for the QRS complex to be all positive in lead aVR — the electrical impulse must be originating from the apex — and impulse origin from the apex virtually assures a ventricular etiology. In My Experience I have not found other QRS morphologies to be helpful in lead aVR. The DOTTED Black-and-White lines in ECG #1 represent my best effort to mark the onset of the QRS complex in all 12 leads in this tracing. I think (it is admittedly hard to be certain) that there istiny initial negative deflection before the positive deflection in lead aVR of ECG #1. (The fact that it is so difficult to determine the direction of the initial deflection in lead aVR of ECG #1 illustrates my reason for ignoring anything but an entirely positive QRS in lead aVR).
MInitial Assessment of ECG #1: I saw a regular WCT at ~190/minute without clear sign of atrial activity. The fact that this patient was middle-aged, and had a known severe cardiomyopathy meant statistically there was a >90% likelihood that this regular WCT rhythm was VT. QRS morphology was a bit atypical for LBBB in ECG #1 (all that notching + slight delay of anterior S wave downslopes) — but given the cardiomyopathy + a QRS morphology that could be consistent with LBBB (upright R wave in lateral leads I and V6 — and predominantly downward QRS in the anterior leads) — I could not rule out the possibility of an SVT rhythm.
  • Initial treatment with Adenosine was completely appropriate. Adenosine is unlikely to be helpful for ischemic VT — but, since QRS morphology in ECG #1 was potentially consistent with an SVT rhythm (with either underlying LBBB or LBBB aberration) — a “therapeutic trial” with Adenosine is exactly what I would have tried in this situation. The ultra-short half-life of Adenosine poses low risk of long-lasting adverse effects — and this drug may convert the rhythm if the etiology was an SVT.
  • If Adenosine is ineffective (as it was in this case) — then synchronized cardioversion becomes the treatment of choice.
  • PEARL #4  Emergency providers often think there are only 2 possible answers for the etiology of a regular WCT rhythm = either definite VT or definite SVT (in which the reason for QRS widening is either preexisting BBB or rate-related aberrant conduction). Remember that there is a 3rd possible answer = which is, that we may not be certain of the etiology of the WCT at this point in time. In this case — I favor a subjective relative probability estimate, based on my best educated guess. I’ve already suggested a ~90% likelihood that ECG #1 represents VT — but 90% is not 100%, and we need to remain open to other possibilities.
  • PEARL #5  There is no need to delay treatment of a regular WCT just because we are not 100% certain of the diagnosis. On the contrary — much (if not most) of the time, we begin arrhythmia treatment of a WCT before we know with 100% certainty what the rhythm is. Both Adenosine and synchronized cardioversion were appropriate initial treatments in this case!

Wwere then Shown ECG #2: I thought the appearance of the baseline tracing on this patient reduced the likelihood that the regular WCT in ECG #1 was VT.
  • As stated above — ECG #2 confirmed that this patient had a baseline wide QRS from complete LBBB + fragmentation in multiple leads.
  • Leads V1 and V2 in the baseline tracing previously showed delay in S wave downslope in this patient with LBBB — so this was not a “new finding” resulting from the wide tachycardia.
  • In addition — I was “struck” by the initial QRS deflection in leads II, III, aVL and aVF — which to my eye had a definite resemblance to the initial deflection in these leads during the WCT in ECG #1. Particularly in leads III and aVF — the resemblance was more than I’d attribute to chance.
  • BOTTOM LINE  After seeing ECG #2, I reduced my estimate that the rhythm in ECG #1 was VT to ~70-80%. This is still more-likely-than-not VT — but I was now less certain than I was before I had seen this patient's baseline tracing. That said, clinically it was time to treat this patient — and a definitive rhythm diagnosis was not essential for initial treatment measures!

That the regular WCT in ECG #1 was not VT was subsequently established — because EP study could not induce VT in the laboratory. Only AFib was induced during EP study.
  • That said — I do not agree that the rhythm strip recorded during Adenosine administration (which I show in Figure-2) proves that the rhythm in ECG #1 was 1:1 AFlutter.
Figure-2: The rhythm strip recorded during Adenosine administration (See text).



Reasons why I do not feel ECG #4 ( = the rhythm strip shown in Figure-2) proves that the rhythm in ECG #1 was 1:1 AFlutter include the following:
  • As previously discussed — the rate of the regular WCT rhythm was ~190/minute. This is different than the rate of flutter waves in ECG #4. By the Every-other-Beat Method — 2 flutter waves (RED numbers in the middle area of ECG #4) are recorded in less than the duration of 3 large boxes on ECG grid paper (BLUE numbers in this middle area in ECG #4). Therefore — HALF the rate for flutter waves is ~110-115/minute — which means that the rate of flutter in ECG #4 is ~220-230/minute, or significantly faster than the rate of the regular WCT.
  • Although difficult to see in ECG #4 because of all the ventricular ectopic beats (beats #7-thru-14 all manifest a different QRS morphology than was seen for beats #1-thru-6 during the regular WCT) — I do not see flutter waves in ECG #4 before beat #14. So rather than Adenosine “revealing” underlying AFlutter that had been conducting 1:1 during the regular WCT — isn’t it possible (if not more probable) that new-onset AFlutter after Adenosine was given may have been caused by Adenosine? (especially since the rate of flutter waves in ECG #4 is at least 30 beats/minute faster than the rate of the regular WCT)?
  • PEARL #6  It is important to be aware of the side effects of Adenosine. These include not only induction of significant bradycardia (albeit usually short-lived) — but also both ventricular and supraventricular tachyarrhythmias. Thus, Adenosine has been shown to induce both AFib and AFlutter — probably because of the effect this drug has on shortening atrial action potentials, and therefore reducing the effective refractory period in atrial tissue (Mallet ML: Proarrhythmic Effects of Adenosine — Emerg Med J 21:408-410, 2004).
  • That said — the fact that QRS morphology for beats #15 and 16 in ECG #4 (during the period of definite AFlutter) is very similar (minus superimposed flutter waves) in almost all leads to QRS morphology during beats #1-thru-6 (during the regular WCT) — strongly supports this QRS morphology as being supraventricular in origin — especially since the PR (or rather flutter-R) interval during AFlutter appears to be constant (the BLUE “PR” in ECG #4— which therefore suggests that flutter waves are conducted. (It would have been helpful to have a longer period of monitoring, in order to be more certain of the above observations ... ).

FINAL THOUGHTs: Our sincere THANKS to Dr. Brooks Walsh for presenting this fascinating case! It emphasizes a series of KEY concepts in emergency arrhythmia diagnosis and management. While statistically — VT was the most likely diagnosis for the rhythm in ECG #1 — a supraventricular etiology could never be discounted. In my opinion, the “rules” were followed — since it simply was never possible to be certain of the rhythm etiology until later during EP testing. Clinically, this did not matter — since appropriate initial treatment (in this case Adenosine, followed by synchronized cardioversion) was the SAME regardless of what the etiology of the regular WCT turned out to be.
  • Even though Adenosine did not convert the regular WCT rhythm — it did support our suspicion that the WCT was an SVT, because QRS morphology during flutter-conducted beats was virtually identical to QRS morphology during the WCT.
  • Availability of a prior (baseline) tracing — and a post-conversion tracing further supported a supraventricular etiology by revealing QRS morphology quite similar (albeit not identical) to QRS morphology during the tachycardia.
  • EP study was needed for confirmation of the final diagnosis.


Recommended Resources