A 48-year-old female arrives at the ED via ambulance after witnesses saw her vomiting large volumes of blood in a local convenience store before collapsing to the floor—obviously, not her best day. She is barely communicating and cannot provide any history. Ambulance personnel have placed two large-bore IVs and started normal saline boluses. Just as you start your initial evaluation, she has a large volume hematemesis. You are concerned about her depressed mental status and the severity of her illness and decide to intubate her. After intubation, the chest wall rises symmetrically and the lungs sound clear. Her heart sounds are distant, and she is tachycardic (pulse 120 bpm). Her blood pressure is 80/40 mm Hg. Her oxygen saturation is 78%.
Question 22.12.1 The standard of care in detecting esophageal intubation is:
Answer 22.12.1 The correct answer is "C." The standard of care is the end-tidal CO2. All of the others are notoriously unreliable. However, they all should be done. If you think you hear breath sounds but the oxygen saturation is not rising and end-tidal CO2 is low, the ET tube is probably in the esophagus.
Question 22.12.2 The end-tidal CO2 can be falsely negative (detecting no CO2) in which of the following situations?
A) Ingestion of carbonated soft drinks (or "pop" as we call it in the Midwest)
B) Intubation in the posterior pharynx above the cords
C) Nasotracheal intubation
Answer 22.12.2 The correct answer is "D." The end-tidal CO2 requires that there is gas exchange in the lungs. If there is no gas exchange, the CO2 will be low. This may be the case during cardiac arrest. "A" can actually cause a false positive. Carbon dioxide in the stomach will give a positive end-tidal CO2 with an esophageal intubation. The same is true of "B." If the patient is breathing spontaneously, the CO2 will be elevated even when the ET tube is above the cords. "C," nasotracheal intubation, should have no effect on end-tidal CO2.
You order laboratory studies that include a CBC, basic metabolic profile, liver chemistries, amylase, lipase, and coagulation studies. In addition, you type and cross for six units of packed red blood cells. The nurses have already contacted the gastroenterologist on call, and she is on her way. You decide to perform nasogastric (NG) lavage.
Question 22.12.3 Which of the following is FALSE regarding NG lavage?
A) NG lavage may be negative even in the presence of an upper GI bleed
B) NG lavage should not be attempted in obtunded patients until they are intubated
C) Iced fluid should not be used to lavage patients with an upper GI bleed
D) The placement of a NG tube is contraindicated in patients who may have variceal bleeding
Answer 22.12.3 The correct answer is "D." Varices are not a contraindication to the use of a NG tube. Studies suggest that a NG tube does not increase bleeding. Both "A" and "B" are true statements. Lavage should not be done in patients who are obtunded or otherwise unable to protect their own airway unless they are intubated. In fact, if the patient has obvious blood in the vomitus, NG lavage is not necessary unless it is to clear stomach contents in order to perform endoscopy or to prevent vomiting. It adds nothing to the management of the patient with an upper GI bleed except confirming the diagnosis. Remember that false-negative NG aspirates occur with intermittent bleeding and bleeding beyond the ligament of Treitz. "C" is also a true statement. Iced lavage fluid should not be used in patients with a GI bleed. The cooler temperature that results from the ice inhibits hemostasis and can increase bleeding.
Contraindications to gastric lavage include known ingestion of hydrocarbons and caustic agents, such as alkalis and acids.
While your nurse is performing the warm water lavage (although why lavage is being done is a separate question since as noted above it does not add anything to this patient's care), you perform a secondary physical examination on the patient. Her most obvious finding is a markedly jaundiced state. Other pertinent findings include a 6-cm scalp laceration, which continues to actively bleed, moderate ascites, and lower extremity edema. After lavage with nearly 5 L of isotonic fluid, the patient's aspirate continues to be bloody. At this point, her vitals are remaining steady but have shown no sign of improvement. Initial emergent labs have returned: Na 142 mEq/L, K 3.2 mEq/L, Cl 106 mEq/L, CO2 18 mEq/L, BUN 40 mg/dL, Cr 0.8 mg/dL, glucose 110 mg/dL, WBC 7,000/mm3, Hgb 9.8 g/dL, Hct 29%, Plt 62,000/mm3, INR 3.0, PTT 48, albumin 2.4 g/dL, AST 76 IU/L, ALT 39 IU/L, Bili 3.5 mg/dL, amylase 210 IU/L, and lipase 24 IU/dL. The gastroenterologist is still 5 to 10 minutes away.
Question 22.12.4 In considering what to do next, which of the following would be most appropriate?
A) Address the presence of platelet dysfunction by transfusing with a 10-pack of platelets
B) In order to accurately assess degree of volume depletion, place a central venous pressure (CVP) catheter or Swan–Ganz catheter, and bladder catheter
C) Regardless of the gastroenterologist, central line placement should be priority at this time because fluid resuscitation is the primary concern
D) Emergent gastric tamponade should be attempted with a Foley catheter
E) None of the above is a great idea right about now
Answer 22.12.4 The correct answer is "E." None of the above is a particularly good idea at this juncture. Looking at them one by one, "A" is not necessary since a platelet count of 62,000 is adequate for hemostasis (although you would not be faulted for giving platelets). A platelet count of <50,000 is considered an indication for platelet transfusion in an actively bleeding patient. "B" and "C" are not good ideas because coagulopathy (INR 3.0) is a relative contraindication to central line placement. It can still be done, but the line is not needed at this point. Recall that peripheral catheters will deliver more fluid more rapidly when compared with central catheters. Thus, two large-bore, peripheral IVs are the access of choice. "D" is incorrect because gastric tamponade with a Foley is like trying to stop a leak in the Hoover dam with putty: it won't work. The most effective treatment at this time may be to give vitamin K and fresh frozen plasma to reverse her coagulopathy.
Question 22.12.5 Effective methods for controlling upper GI variceal bleeding that improve outcomes include all of the following EXCEPT:
C) Transjugular intrahepatic portosystemic shunt (TIPS) procedure
Answer 22.12.5 The correct answer is "D." Vasopressin, while achieving initial control of bleeding in up to 60% of patients, has essentially no effect on rebleeding and no effect on mortality. This may be because ischemia of the splanchnic bed and other areas caused by vasopressin outweighs any benefit. In more bad news, octreotide also does not have any effect on mortality unless combined with variceal ligation. Variceal ligation and sclerotherapy both reduce mortality. Additionally, a TIPS procedure has been shown to effectively stop bleeding by reducing portal pressures. It also reduces acute mortality. Data is conflicting on long-term outcomes, but encephalopathy is more common after TIPS.
An elevated BUN can be indirect evidence of a GI bleed in patients with liver disease. The digestion of blood leads to the elevated BUN.
Despite Dr. Graber's warning to the contrary, you consider placing a central line.
Question 22.12.6 Which of the following locations of central line placement is associated with the highest rate of infection?
D) They all have similar infection rates
Answer 22.12.6 The correct answer is "D." The classic teaching is that femoral lines are more likely to get infected. However, the data doesn't back this up; infection rates seem to be the same for all above locations (Crit Care Med. 2012;40:2479). Subclavian veins have a higher risk of complication such as arterial injury, hemothorax, pneumothorax, and lung injury. Thoracic duct injury is most common with left internal jugular cannulation. Note that ultrasound guidance of central line insertion is the standard of care.
The gastroenterologist arrives and you explain the situation. Vitals at this time include a temperature of 36.8°C, pulse 105 bpm, respirations 14 (ventilator dependent) bpm, blood pressure 85/40 mm Hg, and oxygen saturation of 92%. The gastroenterologist plans to attempt endoscopy with sclerotherapy, but would like to have the general surgeons available for backup in case emergent operative intervention becomes necessary. As you prepare to contact the surgeon, you recall risk stratification for cirrhotic patients is via the Child–Pugh classification system.
Question 22.12.7 Which of the following is FALSE regarding the Child–Pugh scoring system?
A) The Child–Pugh scoring system can be used to predict the risk of variceal bleeding
B) The five criteria used in the Child–Pugh classification are ascites, encephalopathy, albumin, bilirubin, and INR
C) The Child–Pugh scoring system can be used to determine if the patient with a deep hepatic encephalopathy will "wake up"
D) A patient with a serum bilirubin of 3.5 mg/dL may have the exact same Child–Pugh score as a patient with a serum bilirubin of 25 mg/dL
Answer 22.12.7 The correct answer is "C." The Child–Pugh scoring system does not predict the course of hepatic encephalopathy. The other statements are true. The Child–Pugh classification includes evaluation of ascites, encephalopathy status, albumin, bilirubin, and INR. "A" is true because the Child–Pugh score can be used to predict the risk of variceal bleeding as well as the surgical risk and the overall mortality in patients with liver disease. "D" is true since there is a maximum to the Child–Pugh scoring system for each parameter, and the same score is given for all bilirubin levels above 3 mg/dL.
Further discussion: The Child–Pugh classification as a global battery of tests can help to more accurately assess degree of cirrhosis, need for transplant (minimum score of 7), mortality rate from variceal bleed, and outcomes after surgery and TIPS, and outcomes with a hepatoma (Table 22-6).
TABLE 22-6CHILD–PUGH CLASSIFICATION ||Download (.pdf) TABLE 22-6 CHILD–PUGH CLASSIFICATION
|Points assigned ||1 ||2 ||3 |
|Encephalopathy ||None ||Low grade ||High grade |
|Ascites ||None ||Slight ||Moderate-large |
|Bilirubin ||1–2 mg/dL ||2–3 mg/dL ||>3 mg/dL |
|Albumin ||>3.5 g/dL ||2.8–3.5 g/dL ||<2.8 g/dL |
|INR ||<1.7 ||1.8–2.3 ||>2.3 |
Class A is defined as having 5 to 6 points, Class B is 7 to 9 points, and Class C is 10 to 15 points. The 1- and 2-year survival in patients with Class C disease is 45% alive at 1 year and 35% alive at 2 years.
Class A patients have an operative mortality of 1%, while Class B and C have operative mortalities of 3% to 10% and 30% to 50%, respectively. In patients with hepatomas, no Class B or C patients survived 3 to 5 years following resection, although about 40% of Class A patients survive for 5 years.
You quickly recall that there is another prognostic scoring system available for risk stratification and mortality, which is used by United Network for Organ Sharing (UNOS) and called the Model for End-stage Liver Disease (MELD) scoring system.
Question 22.12.8 Which of the following is TRUE regarding the MELD scoring system?
A) The MELD scoring system can be used to predict 3-month survival following TIPS
B) The three criteria used in the MELD classification are bilirubin, INR, and creatinine
C) The primary use of the MELD classification is for assessing mortality risk and prioritizing patients on the waitlist for liver transplant
D) The minimum score is 6
Answer 22.13.8 The correct answer is "E." The MELD score was originally used to predict the 3-month mortality after TIPS ("A"). However, now the MELD score has been validated as a tool to estimate overall mortality related to liver disease and to prioritize patients on the transplant wait list ("C"). The MELD score is computed by a logarithmic formula using creatinine, INR, and bilirubin levels. The formula is quite complex, so thankfully there are handy online calculators readily available (http://www.mayoclinic.org/medical-professionals/model-end-stage-liver-disease/meld-model-unos-modification). For any measured bilirubin, INR, and creatinine less than 1, the formula automatically sets the values to 1 so the minimum score is 6 ("D"). The MELD score is interpreted as follows: score ≥40, 71.3% mortality in 3 months; 30–39, 52.6% mortality in 3 months; 20–29, 19.6% mortality in 3 months; 10–19, 6.0% mortality in 3 months; <9, 1.9% mortality in 3 months.
Objectives: Did you learn to…
Manage a patient with a massive upper GI bleed?
Determine if an ET tube has been placed correctly?
Recognize the uses and limitations of gastric lavage?
Recognize the uses and limitations of central line placement and Swan–Ganz catheter placement?
Evaluate a patient with liver disease, using the Child–Pugh and MELD classification systems?