NG tubes. just. wont. die.

My angst for the NGT has been explained in a previous post, and while this study adds to said angst, it sadly comes short of putting a nail in the coffin in the debate with surgical colleagues.
This is a retrospective single center study which enrolled 181 ED patients with SBO from September 2013 to Sept 2015, and essentially grouped patients according to whether or not a nasogastric tube was placed (49% of patients did not receive the dreaded NGT). Looking at a multitude of factors, they attempted to tease out items associated with nasogastric tube placement, and if there were any appreciable benefits to NGT placement.

Ultimately, if you are over age 70 (37% NGT+ vs 19% NGT-),  have a malignancy (30% NGT+ vs 17% NGT-), or had a prior SBO (56% NGT+ vs 32% NGT-) you’re more likely to have an NGT because, hey, one good NGT deserves another.  NGT+ patients were also less likely to have “likely / early SBO” (19% NGT+ vs 40% NGT-) on CT imaging as well.

All in all, while I’d love to point at the mean length of stays (7 days for NGT+ vs 4.2 days for NGT-; median 5 days vs 3 days), and non-statistically significant resection rates of 13% vs 9% as indications that the NGT is not needed…. well, we’re not exactly comparing apples to apples. The NGT+ patients were sicker- they were older, had higher malignancy rates, had a slightly higher surgical rate, and were more likely to have “definite SBO” on CT. Sadly, this is not the paper to put the NGT argument to rest.  We still need a larger study, preferably with matched controls, to fully put this dinosaur to rest.

 

Someone?  please? … anyone? please?

Versed, Pavlov, & the NGT

I admire & strive for efficiency, empathy, and efficacy.  Thus, if a certain painful procedure extends lengths of stay without significant benefit to the patient, I’m hesitant to order it. However, sometimes it is frowned upon to not play nice in the sandbox with our Pavlov consultants who request NG tubes for small bowel obstructions.

 

Thus, if you must place an NG tube for an obstruction, keep today’s study in mind.

Up at The U (of Vermont), patients that were ordered an NG tube and were between 18 and 60 years of age were given 1 ml of intranasal atomized co-phenylcaine (lidocaine 5% with 0.5% phenylephrine to each nostril, followed by either placebo or 2 mg of IV midazolam with NG tube placement within 5 minutes of placebo or midazolam. Using a 100 mm VAS for pain and discomfort, as the primary endpoint, the patients were interviewed 15 or 45 minutes after the procedure.

After 51 total cases (23 of which qualified for the study), the trial was stopped early as several ED clinicians felt that midazolam prior to insertion was superior to topical anesthetics alone and did not want their patients randomized in the study.

Hence, the numbers are a good bit low. Only 13 patients in the control arm, and 10 in the treatment arm. The mean difference in pain scores was 31mm and 36mm for discomfort. None of the midazolam placements were rated difficult by nursing who placed the NG tube, whereas 3/13 in the control group were rated as difficult or very difficult to place.

So yes. The numbers are small, and the authors shed some light on questions we would have. They spoke about anecdotally about 1mg of midazolam for those over 60 years old or with known pulmonary disease – which they felt did not alleviate pain or discomfort. Here, I think a 1mg bolus followed by 0.5mg boluses as needed is reasonable.

They cite a paper stating that 91% of providers would change their practice if new literature showed a convenient way to reduce patient discomfort. Well, here it is. So, if Pavlov’s dog barks and you HAVE to place an NG tube, hopefully your institutional policy allows IV midazolam without considering it procedural sedation (or, potentially using intranasal midazolam for anxiolysis).

Increased sono, decreased negative peds appy rate.

Academia and community ED settings are very different, and sometimes one community to the next is often times very different as well. This study– which included 35 pediatric ED’s and 52,000+ cases of appendicitis demonstrates this. Essentially, the authors took all their appendicitis data from 2010-2013, and looked at what imaging, if any, the patient received, negative appendectomy rates, length of stay, frequency of perforation, and 3 day bounceback rates from initial ED visit (ie, did ultrasound or CT “miss” an eventual appendicitis?)

Ultrasound use increased from 24% in 2010 to 35.3% in 2013 for the sole diagnostic modality in what eventually was diagnosed as appendicitis. CT rate decreased 21.4% in 2010 to 11.6% in 2013. Almost 50% of the patients (25,254 out of 52,153) had neither CT or US (!). The negative appendectomy rate in 2010 was 4.7%, while in 2013, it was 3.6%. There was no difference in length of stay, frequency of perforation, or 3 day bounce back rates from 2010 to 2013.

Ok. If you work in a community ED, raise your hand if 50% of your potential appendicitis cases get seen by surgery without any imaging. Only 11.6 % CT use? I think most community surgeons would be less than thrilled to be called at 2am for a “rule out” appendicitis evaluation without imaging.

While I think the authors live in an alternate universe, I think this paper makes a great argument to do the right thing. It would be great if non-academic settings could utilize US and surgical consultations for a “hot belly” without shipping them off to radiology. Or better yet, have surgery perform POCUS with you to confirm findings.

Who is really distracted by distracting injuries?

Those that (still) recommend the pan-scan in trauma, it would seem.  Over a once year period, this 803 patient prospective study (451 of which with distracting injuries) evaluated all awake, alert, blunt trauma patients with a GCS of 14-15 to determine the validity of an abdominal exam.  Endpoints were injuries which required the OR and those which required a transfusion.

A ten percent failure rate of the abdominal exam for an intra-abdominal injury seems rather high, but this is ten percent of all who actually had an injury. When you consider “all-comers,” it’s more like 1.1% of those with a distracting injury and 0.57% for those without a distracting injury.  With that, you would need to scan over a hundred patients to find one intra-abdominal injury that you would have otherwise missed.  Throw in the fact that all five missed injuries in the distracted cohort had solid organ injuries, and that none of them required surgical intervention or blood transfusion, and you can see how the authors come to their conclusion:

“These data suggest that clinical examination of the abdomen is valid in awake and alert blunt trauma patients, regardless of the presence of other injuries.”

$ISRG – great marketing, average results.

Doctors often recommend surgery when medicine and lifestyle changes cannot ease your abdominal or gastrointestinal symptoms. If your doctor recommends surgery, learning about all surgical options can help you to make the best decision for your situation.

Surgery is often an effective treatment for many abdominal and GI conditions, but traditional open surgery with a large incision is highly invasive. Open surgery may also require a long hospital stay and lengthy recovery.

Fortunately, there are minimally invasive surgical options. The most common is laparoscopic surgery (laparoscopy). Surgeons make a few small incisions and insert a tiny camera and long-handled instruments to reach inside your abdomen. The camera transmits images onto a video monitor in the operating room to guide surgeons as they operate. Traditional laparoscopy is effective for many routine procedures but the rigid surgical instruments can be technically challenging for delicate or complex operations.

da Vinci Surgery is another minimally invasive option for patients facing abdominal or gastrointestinal surgery. Instead of a large abdominal incision used in open surgery, da Vinci surgeons make a few small incisions – similar to traditional laparoscopy. However, the da Vinci System features a magnified 3D high-definition vision system and special wristed instruments that bend and rotate far greater than the human wrist. As a result, da Vinci enables your surgeon to operate with enhanced vision, precision, dexterity and control.

State-of-the-art da Vinci uses the latest in surgical and robotics technologies and is beneficial for performing complex surgery. Your surgeon is 100% in control of the da Vinci System, which translates his or her hand movements into smaller, more precise movements of tiny instruments inside your body.

from http://www.davincisurgery.com/da-vinci-general-surgery/

Sounds great! Less scarring, I make it easier for my surgeon than having to use those bulky tools, and I’ll decrease my hospital stay and recovery! Its a robot that my surgeon is “100% in control of… for more precise movements!” Sign me up!

Then, you see (yet another) study like this…

Today’s discussion is on a multicenter study comparing 223 robotic performed gastrectomies vs 221 laparoscopic gastrectomies, where patients were matched according to the surgeon, extent of gastric resection, and sex, and the primary outcome was morbidity and mortality. Overall complication rates were similar (11.9% for robotics vs 10.3%), which major complications being 1.1% for both groups. Length of procedure was significantly longer for the robotic group (221 minutes vs 178 minutes), with, obviously, significantly higher costs in the robotic arm ($13,432 vs $8090 USD). There was no significant difference between estimate blood loss, rate of conversion to open procedure, diet build-up, or length of hospital stay.

In case you were wondering, as of June 4, 2015, Intuitive Surgical- the maker of the da Vinci – has a market cap of 18 billion dollars, while comparatively, TeamHealth and Envision (ie, EMCare) – the two largest EM staffing companies in the US, are valued at 4 billion and 7 billion, respectively.

At over $5,000 more per procedure without any obvious benefit, why is the da Vinci even an option?

Keep those toes warm with tPA!

A graduate student decided to go out and celebrate after doing well on an exam.  After being unable to locate a designated driver, and having their iPhone run out of battery rendering them without Uber to locate a taxi, they decided to sleep in their car for a few hours.  Unfortunately, it was bitterly cold out that evening.  After waking up several hours later and returning home, the patient was unable to rewarm their feet despite earnest attempts.  They began to notice blisters on their feet, and then proceeded to the nearest Emergency Department for further care, where the astute provider diagnosed the patient with advanced frostbite and began making phone calls.

Are there any non-surgical interventions that may help?

In a 2005 publication, 19 patients with severe frostbite received tPA (6 of which received it intra-arterially) in addition to heparin.  These patients had not improved with rapid rewarming and had absent doppler pulses in the distal limb or digits, and no perfusion by technetium scan.  Historical data suggested that 174 digits in these 19 patients were at risk for amputation, yet only 33 were amputated.  Poor response was noted in those with >24hrs of cold exposure, warm ischemia >6hrs, or evidence of multiple freeze/thaw cycles. 

 

In a 2007 publication of a single center study, patients with frostbite from 1995-2006 that were not treated with tPA were compared to patients who were treated with tPA from 2001-2006. 

 

The data:

6 tPA treated patients, encompassing 13 extremities
26 non tPA treated patients , encompassing 57 extremities.

Percentage of extremities requiring an amputation
tPA treated patients – 23%
non-tPA treated patients 55 %

Percentage of patients requiring a digital amputation:
tPA treated patients – 23%
non-tPA treated patients – 29%

Percentage of patients requiring a proximal amputation:
tPA treated patients – 0%
non-tPA treated patients 25% (about a third of which were below knee amputations)

one of the tPA treated patients developed a retroperitoneal hematoma that resolved spontaneously without intervention. 

While these are single center studies, and not truly blinded, I think that they provide a reasonable basis to attempt tPA if you have an IR capable institute – or make a reasonable argument to transfer to an institute that does.  There seems to be at least a 50% decrease in the number of surgical procedures performed  and a decrease in the size of area to be amputated.  It is certainly worth mentioning to the admitting surgical team.  Drop the bug in their ear, and the data suggests a strong chance that you’ll save a limb.

Stay warm out there.  

 

 

PMID: 16394908
PMID: 17576891

Surprise! Admitting to surgery for a surgical problem is better for everyone!

A chart review of patients who underwent a cholecystectomy for mild gallstone pancreatitis was performed on 50 patients with mild gallstone pancreatitis (As per Atlanta Classification) that were admitted to medicine and 52 that were admitted to surgery. Moderate to severe gallstone pancreatitis and non-gallstone pancreatitis were excluded. Medical admissions were more likely to be older and have more comorbidities. Surgical patients had a shorter time to surgery (44 vs 80 hours; P < .001), a shorter LOS (3 vs 5 days; P < .001), and significantly lower hospital costs. Subgroup analysis was performed on patients with an ASA score of one or two which matched well. All outcomes still favored surgical admission.

Sure, this is a chart review based on the patients who had a cholecystectomy, and does not include the number of patients that were solely medically managed, and does not include patients treated with ERCP without cholecystectomy. A Cochrane Review found that early ERCP for patients with biliary obstruction was associated with a non-significant trend towards reduction of local and systemic complications as defined by Atlanta Classification. Regardless, this should likely be a conversation between GI and general surgery, and not between GI, general surgery, and medicine.

I have recently wondered if there was any literature to support specialty findings (ie, gall stone pancreatitis, hip fractures) should go to a specialty service, rather than all findings going to medicine. In this particular case, there is a shorter time to surgery, shorter length of stay, and lower cost without an increase in patient morbidity or mortality. Would these findings cross over to orthopaedics for hip fractures? What about renal stones to urology? Syncope to cardiology?

.

.

.

PMID: 25312841
http://pubs.rsna.org/doi/full/10.1148/radiol.11110947

SBO, NGTs, and Primum non Nocere

You just got the call from radiology that your pleasant, frail 88 year old patient has an obstruction secondary to adhesions. There is no evidence of mesenteric ischemia on CT. The patient has a normal lactate, normal wbc, is afebrile, and is not peritoneal. He vomited once on arrival 3 hours ago, but is no longer nauseated after a one time dose of 4mg Zofran. You cringe for your patient knowing what surgery is going to say when you page them.

But whats the evidence for the infamous NG tube?

.
.
.
.
.
.
.
.
.
.

(the above lines left blank intentionally)

Even the guidelines for management of SBOs can not provide links for an NGT vs no NGT trial. Level of evidence for NGT? 2b. Grade of recommendation? C. “early tube decompression… may be beneficial in the initial management of non-strangulating (acute small bowel obstruction), in adjunct with fluid resuscitation and electrolytes imbalances correction.” (italics & underline placed by this blogs author)

So what is the evidence?
290 patients with SBO, NGT vs no NGT:
pneumonia and respiratory failure was significantly associated with NGT placement. Time to resolution and hospital length of stay were significantly higher in those with NGTs.

What else can you do?
Study of 128 patients with partial adhesive SBO:
IV hydration / NGT / NPO vs IV hydration / NGT / magnesium oxide PO / lactobacillus / simethicone:
76% vs 91% successful treatment without surgery. 4.2 vs 1.0 days in hospital. Similar complication and recurrence rates. (level of evidence 1b, Grade of Recommendation: A)

Study of 70 postoperative SBO with obliterative peritonitis:
TPN / NGT vs TPN / NGT / somatostatin/ dexamethasone:
29.9 vs 22.2 days to resolution. 16.7 days vs 9.9 days of NGT use. 25.8 day vs 34.9 day length of stay. Complication and relapse rates comparable.

Unless patients are profusely vomiting despite multiple rounds of anti-emetics, I try to avoid placement of NGTs. Its a miserable procedure that leads to worse outcomes. First, do no harm.

.

.

.

Nonsurgical management of partial adhesive small-bowel obstruction with oral therapy: a randomized controlled trial. PMID: 16275967
Conservative treatment of early postoperative small bowel obstruction with obliterative peritonitis. PMID: 24379592

Routine nasogastric decompression in small bowel obstruction: is it really necessary? PMID: 23574854

Bologna guidelines for diagnosis and management of adhesive small bowel obstruction (ASBO): 2013 update of the evidence-based guidelines from the world society of emergency society ASBO working group. PMID: 24112637