Thursday, November 14, 2013

1. Most people have 500-600 cm of small bowel, you need at least 150 to prevent short-gut syndrome. If you have to resect beyond that (for strictures in IBD or perforated small bowel diverticulitis, say) then people will have to be on TPN. When you resect a portion of small bowel, be careful not to cut at the root of the mesentery-- if you take out the SMA by accident, the entire small bowel will go necrotic.
2. Principles of diverticulitis management: 
-Bowel rest, antibiotics, fluids. If someone is relatively healthy, you can have them stick to a liquid diet and take PO antibiotics at home (broad spectrum, 7-10 days), and then switch to a high-fiber diet later on. If they're old or sickly, you'll want to admit them, make them NPO, put in an NG, give IV fluids and antibiotics. If they get another event (30% chance), then elective resection is recommended 4-6 weeks after the inflammation dies down. Usually it affects a small segment of sigmoid colon, sparing the rectum and the proximal bowel.
3. Massive lower GI bleed-- likely vascular ectasia (AVMs) or diverticulosis. Other rarer explanations are aortoenteric fistula (suspect if they've had any major vascular surgery or stents in aorta or iliacs that could have eroded into bowel), varices, meckel's, polyps, IBD, hemorrhoids. Cancer is unlikely to cause massive bleeding. Stabilize first. Try to quickly find the source of bleeding with an NG lavage to r/o upper GI, and an anoscope to r/o hemorroids and varices. If the bleed stops (and 90% of AVMs and most diverticulosis cases will stop bleeding spontaneously, although there's a 20-25% risk of rebleed), then colonoscopy at a later date. If they will not stop bleeding and are crashing, do not attempt colonoscopy-- you won't be able to see, and you risk perforating. Instead, do tagged RBC scan (better for more stable patients) or angiogram (better for sicker patients), find the source, and then go to the OR for hemicolectomy. If you dont know where it's coming from, you may need to do a total colectomy. Most surgeons will wait to go to the OR until the person has gone through 4-6 units of transfusions of pRBCs; exceptions-- people who you don't want to transfuse, like Jehovas' witnesses or people with lots of antibodies or an odd blood type, people who are acutely highly unstable.
4. Sigmoid volvulus can be decompressed with a rigid sigmoidoscopy. Cecal volvulus is a surgical emergency, attempts to decompress with barium enema or scoping do more harm than good.
5. Ogilvie's-- dilation happens preferentially in the cecum and R colon because of the law of laplace. Try neostigmine first. Manage with surgery (R hemicolectomy) if the diameter exceeds 11-12 cm in an immunocompetent person, less in immunosuppressed.
6. For squamous cell anal cancer: very small, superficial lesions with no nodes can be superfically excised. Anything larger should not get resection first; it should get the Nigro protocol, and then surgery only if there is residual cancer afterwards. 
Nigro protocol:
-30 gy to the tumor, 2 gy/day for 5 days a week for 3 weeks (days 1-21) 
-5-FU, 1000mg/m2/day, continuous for 4 starting on day 1 of radiation, then another 4 days from days 28-30. 
-mitomycin-C 15mg/m2 IV bolus on day 1. 
This protocol has proven very effective, even in big cancers with +nodes, frequently able to lead to complete control without surgery, allowing people to avoid the major morbidities associated with abdominoperitoneal resections. {review 1} {review 2}
7. EC fistulas revisited: 
-High output (>500mL in 24 hrs) must be managed with total bowel rest: NPO, TPN.
-Low output, people can be allowed to eat.
-The further away the fistula is down the small bowel, the less likely it is to compromise nutrition and the more likely it is that things will be absorbed by the bowel proximal to the lesion and it'll be low output.
-EC fistula of the colon are much higher risk for infection
-Prophylactic antibiotics only if there is an indication-- increased risk of infection ie metastatic cancer that is obstructing distally and putting back-pressure on fistula, decreasing healing
-Wait at least 6-8 weeks to operate. 75% of EC fistulas will heal on their own. If you go in right away, and there is inflammation and infection, you risk making things worse as the bowel will just fall apart rather than healing strong.
-Principles of care while waiting for it to heal: sepsis control (drain any abscesses), nutrition, octreotide/NG/bowel rest.
8. Things that make EC fistulas less likely to seal on their own:
-Foreign body
-Radiation history
-Infection, inflammation (crohn's for example)
-Epithelialization
-Neoplasm,
-Distal obstruction
9. Pancreatic cancer: 
-Overall mortality, all-comers, for a whipple procedure: around 8%, <3% at high-volume centers. Risk tacks up if people are sick. Needing to do a portal vein reconstruction doubles your risk.
-Solid components and wall nodules in a cyst are a bad prognosis
-Vascular invasion is a bad prognosis-- watch out for portal vein, SMA/SMV, celiac a
-Neuro invasion is a bad prognosis-- presents as a constant, gnawing, non-radiating pain that feels like a hot poker in your back. Versus radiculopathy or mechanical back pain which radiates and changes with position. Back pain from pancreatic cancer represents invasion into sympathetic plexus
-CA-19-9 is normally 37, higher numbers represent increased likelihood of metastatic disease.
-Erlotinib plus gemcitabine are more effective than gemcitabine alone, increases mean OS 14 days....
10. Most insurance companies will not pay for epo if your hb is over 11; they will titrate you to 11 but then not beyond that, even if you are facing a surgery which could entail a significant blood loss

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.