Clinical Guide The Nerd's Guide to Pre-Rounding Table of Contents Appendix 1. Tasty Bits: Good Things to Know Up Front Tips on Pharmacology You don’t have to know the dose and schedule of every drug in the PDR to seem well-informed about drug topics on the wards and clinics. Especially on inpatient wards, a relatively small set of drugs get used over and over, and you would do well to make yourself especially familiar with them. They are: anticoagulants (warfarin, heparin, enoxaparin); pain meds (esp. opiates); insulin; and bowel regimens (docusate, bisacodyl, Sennokot, and similar). Bowel drugs. We didn’t study them much in pharmacology, but they are handed out like candy on the wards. For a decent overview, see Prather CM, "Evaluation and treatment of constipation and fecal impaction in adults," Mayo Clin Proc 1998;73:881-7. Key point: If you write a scrip for opiates, your next act should be an order for docusate (Colace). Sliding scale insulin (SSI). This is an example of how insulin orders are commonly written on the wards: FSBS q 4 hrs Insulin, regular, human. Sliding scale, given SQ following FSBS[20] level: 0-80, give juice and call HO[21]; 81-200 = 0 units; 201-250 = 4 units; 251-300 = 6 units; 301-350 = 8 units; 351-400 = 10 units; >400, give 12 units and call HO We did not have any formal reading on "sliding scale insulin" in pharmacology. There’s a reason. It’s not evidence-based medicine, but it’s been around for decades. SSI regimens have the defect of giving insulin after high blood sugar levels appear, and not giving any when sugars are normal even if the patient is taking in food destined to raise them. Patients also have to get stuck round-the-clock, even in the middle of the night when they’re not eating. SSI is used mainly because it’s simple and convenient for house staff. Several studies on SSI demonstrate its problems. In one, patients on proactive insulin regimens stayed in the hospital for less time and had lower sugar levels than did SSI patients.[22] In another study, patients on SSI had a three-fold higher risk of hyperglycemia compared to patients on no drug regimen at all![23] For more information on the efficacy of SSI and other "medical myths," see Paauw DS, "Did we learn evidence-based medicine in medical school? Some common medical mythology," JABFP, Mar-Apr 1999;12:143-9. Acid Base Algorithm When analyzing an ABG (arterial blood gas) result… Look at the pH. Acidemic, or alkalemic?[24] Is the primary problem metabolic, or respiratory? The PaCO2 will tell you. Is the patient compensating for the problem? Check the HCO3. If adjusting acutely, the pH changes 0.8 for every 10 of the PaCO2. If patient has had time to adjust, it’s 0.3 pH/10 pts PaCO2 Calculate the anion gap: Na – Cl – HCO3 = gap (8-12=normal, no gap). Low albumin makes gap artificially low; add 2 to the gap for each point of albumin below normal. If gap is >20, patient certainly has a gap acidosis Calculate the "delta gap": the measured gap – 12 =delta gap, also called the "gap-gap" Add the delta gap to the HCO3: If >30, patient also has a metabolic alkalosis (beyond whatever else is going on). If <23, patient has a metabolic, non-gap acidosis (again, in addition to other issues) Special note: An anion gap of 12 ("normal") may indicate acidosis in multiple myeloma "MUDPILERS" Methanol, Uremia, DKA, Para-formaldehyde, Iron/Isoniazid, Lactic acidosis, Ethanol/Ethylene glycol (antifreeze), Rhabdomyolysis, Salicylates (also Sepsis—which is really lactic acidosis) Metabolic alkalosis: If urine chloride is low: vomiting/NG tube; past diuretic use; post-hypercapnea. If urine chloride is normal or high: excess mineralocorticoids (Cushing’s, Conn’s); current or recent diuretic use; excess alkali administration Metabolic, non-gap acidosis: i.e. excessive loss of HCO3: RTA (Renal tubular acidosis), or RTA ("Rectal tubular acidosis", i.e. diarrhea) (For more detail, see excellent article by Haber R, "A practical approach to acid-based disorders," West J Med 1991;155:146-51.) Approach to the Chest X-Ray You should be taught how to read chest x-rays on your medicine clerkship. However, you may not get to medicine until the end of the year, and you may get pimped on reading a chest x-ray sometime before then. As with most procedures at this stage of the game, I think most of your superiors will be satisfied if you demonstrate you have a METHOD for approaching a task, rather than your simply getting the answer. If someone holds up a CXR film and asks you to read it, I suggest you explain what you're doing as you go. Example: "Well, let's see. The tag indicates this is a film of patient John Smith. It's a chest film… looks a little fuzzy, and the heart's a little big, which makes me think it's AP… and the tag indicates it's an AP film, anyway, so that confirms it. The inspiration looks like it's down to about rib 10, I can see the spinous processes, clavicles look even, so, good quality film. And… I don't see any obvious bone or soft tissue lesions. I see the gastric bubble below the diaphragm, the mediastinum is centered and not enlarged, I see the cardiac borders clearly and the silhouette is not enlarged, and the lung fields appear clear, no densities. Normal chest." By "milestoning" your read, you let your superior see that you have a systematic way of going about it. Below is a sytematic approach to the chest x-ray: Confirm the name of the patient. Evaluate film quality. The mnemonic is "PIER": Position, Inspiration, Exposure, Rotation. Position: Typically, upright PA and lateral. Sick patients will have the fuzzier supine AP (because the film is slid under their chest as they are lying down). Inspiration: Lung fields should extend to about 10th or 11th rib. Exposure: If the film is penetrated enough, you should be able to make out the spinous processes "inside" the vertebrae. If the film is underexposed/too white, you won't be able to see them. If the film is overexposed/too black, bony details will be lost. Rotation: Space between the medial clavicle and the margin of the adjacent vertebra should be equal on the right and left sides. Check the bones. Scapula, humerus, shoulder joint, clavicle, vertebrae, ribs. Look for lytic lesions, new or old fractures, symmetry, osteoporosis, scolisis, rib notching. Check soft tissue. Breast shadows, supraclavicular regions, axillae, chest wall. Look for thickness, subcutaneous emphysema (air bubbles-dark spots), calcifications (bright spots). Diaphragm. Outline should be smooth curves taking off from the midline at 10th-11th rib. Locate gastric air bubble. Look for pneumoperitoneum. Mediastinum. Check trachea-should be to right of midline as it approaches carina. Note the great vessels, esp. aortic arch. Look for adenopathy, mass lesions. Cardiac silhouette. Check size and shape of heart. Right border should be a finger width to the right of spine. Left border should be distinguishable. Heart width should be less than half of the widest chest diameter. Lungs. Check one field, then the other, then compare. Is vasculature engorged? What about the bronchial tree? CP angles visible? Pneumothorax (check very carefully at the apices)? Pleural thickening? Inflitrates? Cavitation? When describing what you're seeing in the lung fields of a CXR, describe the pattern, but don't diagnose. Don't say "Looks like fluid in the fields." Say, "I see patchy densities bilaterally, with a silhouette sign obscuring the left cardiac border. This is an alveolar pattern that could be due to pus, such as from pneumonia, or from fluid due to pulmonary edema." Types of lung densities: alveolar: patchy, poorly marginated. Represents material other than air in the airspaces. May see "air bronchograms"-black lines representing air-filled bronchi amidst water-density alveoli. May note "silhouette signs"-organs' margins blurred by dense material in alveoli of nearby lung tissue. interstitial: thickening of bronchi, septae. Linear or finely granular patterns of abnormal shadows. "Kerley's B-lines" (not "curly" B-lines) are thickening of interlobular septae and are small, bright, horizontal lines seen esp. towards the bases of the lungs. They are associated with CHF. The interstitial pattern is seen in CHF, interstitial fibrosis, cancer, inflammation. atelectasis: loss of volume leads to a shift of interlobar fissures & mediastinum towards the collapsed region. nodules: one or more dense, bright, round lesions. Adenoma, granuloma, cancer, cyst, lymph node, etc. other: abscess (lucency within density, air-fluid level); pneumatoceles (air-containing spaces seen with some pneumonias); honeycombing (airspaces w/thick septae). The above is adapted from notes by Dr. Hugo Yang, Chief Resident, Mt. Zion, autumn 1999, as well as teaching materials from Dr. Marcia McCowan of the Department of Radiology, VAMC. Approach to the EKG On the wards, no one is going to expect you to diagnose Wolff-Parkinson-White syndrome on your first EKG read. As with the CXR read, they want you to show that you have a METHOD for doing the read. Dr. Tom Evans has written a nifty little set of "cribsheets" for analyzing EKGs, but his book is a little advanced for novices. However, getting used to the "Litany of Categorization" outlined below will allow you to eventually take advantage of the Evans method. This basic approach is taken from a talk by Dr. Paul Varosy, Chief Resident, Mt. Zion, 1999-2000. Recall the basic shape of the EKG wave (P, QRS, and T waves). Recall that the EKG paper has large boxes, subdivided into 5 little boxes. Five big boxes is 1.0 seconds. One big box is 0.2 sec. One little box is 0.04 sec. As you look at the EKG, focus on a familiar shaped curve (I usually use lead 2) and answer the following questions: What's the rate: slow, normal, or fast?Usually printed at the top of the EKG printout. You can also count the number of QRS complexes per large box and estimate. Since each box is 0.2 sec, one QRS per box is about 300 beats per minute (bpm). One complex per two boxes is about 150, one complex per three boxes is 100, and so on. The mantra is: 300-150-100-75-60. Any rate under 60 is "bradycardia." 60-100 is normal. Over 100 is "tachycardia." Is the QRS complex wide, or narrow? In other words, is it over .12 seconds, or under. Again, the width of the QRS is often printed at the top of the EKG sheet, or you can inspect it directly (since each "little box" is 0.04 sec, a QRS complex should not be any more than three little boxes). Narrow QRS complexes indicate impulses are travelling through the normal cardiac electrical circuits. Wide QRS complexes indicate that impulses are travelling through ectopic channels. Is the rhythm regular, or irregular? In other words, are there normal, consistently-shaped P waves? Is every P wave followed by only one QRS, and does each QRS get followed by only one P wave? That tells you if you have "normal sinus rhythm." You can also check to see if the length between R groups (the R-R length) is always the same (regular), or if it changes. If it's regular, is it a normal sinus rhythm? If so, the P waves should be erect (look like little hills) in lead 2, and "biphasic" (S-shaped) in lead V-1. If it's irregular (i.e. if the R-R length changes over time), is it predictable, or is it "irregularly irregular"? Irregularly irregular is usually an indication of atrial fibrillation. At this point, you should be able to use the Evans book to identify a set of likely EKG diagnoses (see Evans, p. 34). The rest of the litany, below, is for completeness. Is the heart's axis deviated? I'm not going to get into the relationship between heart position and axis, the various leads, and how/why a shift in the heart's position changes the curve shape on various leads. Look in your physiology books and the Evans book if you want the background. Below is a quick way to assess if the axis is right- or left-shifted. Look at the QRS complex in Lead 1 and Lead AVF. (In the old-style printouts, these were easy to compare, since the curve from Lead 1 sat right above AVF.) If the complex from 1 points down, and the complex from AVF points up, they are pointing towards, or returning to, each other. Returning = Right-shifted. If the complex in 1 points up, and the complex in AVF points down, they are pointing away from each other, or leaving each other. Leaving = Left-shifted. Examine the intervals. Is the PR interval less than 0.2 sec (one big box)? PR intervals > 0.2 sec = heart block. Long PR intervals but otherwise normal trace, with no missing QRS complexes indicates a 1st degree HB. Long PR intervals with some missing QRS complexes indicates a 2nd degree HB. If PR intervals progressively lengthen before the missed QRS= 2nd degree HB, Type 1 a.k.a "Wenckebach". If PR intervals don't change before missed QRS = 2nd degree HB, Type 2. If P and QRS are completely dissociated, think 3rd degree/complete HB. QT interval. As a rule, the QT should be < ½ the RR interval. Look for bundle-branch block: Look at leads V-1 and V-2. Right BBB is indicated by QRS complexes that show two symmetrical peaks ("Rabbit ears" = Right BBB). Left BBB is indicated by a deep S wave. See the Evans book for more detail. Look for signs of ischemia. T wave flattening or inversion is usually the first change seen. ST segment depression indicates a subendocardial injury pattern. ST segment elevation ("tomb-stoning") indicates an infarct. With Q waves can indicate an evolving infarct. Without Q waves indicates acute myocardial injury. Note that "Q-wave" and "Non-Q-wave" MI's are treated differently. Again, see the Evans book for more explanation, and diagrams. Cardiac Tests You May Not Have Heard Of One thing that annoyed me on the wards was references to cardiac diagnostic tests that I had never heard of. You will hear your residents throw around terms like "Patient had a P-thall in October that showed no abnormalities in uptake," or "P-MIBI results consistent with 80% constriction in the LAD." Excuse me? What's a P-MIBI? Or P-thall? These terms refer to imaging tests to assess coronary artery disease. The tests evaluate the flow of blood through heart tissue. To do them, a patient is either put on a treadmill to exercise-a "stress" test-or is given an infusion of a drug to dilate the vessels chemically, which is a "non-stress" test. The latter type of test can be done using the drug dipyridamole (brand name Persantine), or using adenosine. Once blood is flowing, you have to measure it. This is done with scintigraphy-i.e. nuclear imaging tests using radioisotopes. The radioisotopes used are thallous chloride (Tl-201) or technetium (Tc 99m), which has a short half-life that is extended by linking it to Sestamibi. Thallium is a potassium analog and is taken up by cells at a rate proportional to the bloodflow. Technetium is a calcium analog and, when hooked with Sestamibi, is also taken up at a rate related to flow. Thus, a "P-MIBI" is a non-stress test that induces vasodilation with Persantine and assesses flow using technetium-Sestamibi as the imaging agent. A "P-thall" is similar but uses thallium as the imaging agent. In case there are Godzilla-nerds among you for whom the above was not enough information (God help you), see Chou TM et al., "Evaluating CAD Non-invasively: Which Test, for Whom?" West J Med 1994;161:173-80. Fluid Basics From day 1, you will be dealing with fluid management decisions in your ward patients. We didn't get much practical instruction on fluid management during the first two years. A lot of unfamiliar terms may be thrown around in the first few weeks, which can be intimidating. The notes below are intended to give a basic framework and vocabulary. The types of fluids used: Commonly used fluids are bolded in the following table. Values are milliEquivalents per Liter (mEq/L). "D5" refers to 5% dextrose solution. "D5W" indicates dextrose in water. Adapted from Taylor DS, "Fluid and Electrolytes," in Lyerly HK (ed.). The Handbook of Surgical Intensive Care, 2nd edition. Chicago: Year Book Medical Publishing, 1989: 230. Lactated Ringer's is a maintenance fluid favored by surgeons, traditionally. The lactate is quickly converted to HCO3 by the liver, then converted to CO2 in the blood. Thus, not a good choice for a patient who's accumulated high levels of CO2. On medicine, D5 ½ NS is the typical IVF. By the way: dextrose is not added to IVF for significant nutritional purposes, or to prevent muscle breakdown. It's added to keep the Krebs cycle going, to prevent ketoacidosis. Fluid Na Cl K HCO3 Ca osm kCal/L Crystalloids Normal Saline (NS) 154 154       292   D5NS 154 154       565 200 ½ NS 77 77       146   D5 ½ NS 77 77       420 200 D5 ¼ NS 34 34       330 200 D5W           274 200 D10W           548 400 Lactated Ringer's (LR) 130 104 4 28 3 277   3% NaCl 513 513       960   Colloids HESPAN 154 154       310   Plasminate (5% protein) 145 100 0.25     310   25% albumin 130-160 130-160 1.0     310   How to determine which fluid to use, and at which rate: First, you must decide if the patient is on maintenance or resuscitation. Maintenance means replacing fluids lost from normal physiological functioning. Resuscitation means pt has past or ongoing fluid losses that need to be replaced, i.e. hemorrhage, diarrhea, etc. Maintenance Fluids. Best maintenance fluid for adults: D5 ¼ NS + 10-20 mEqs KCl (note D5 ½ NS is often used, but that will give 185 mEqs of excess Na per day; not a big deal if you have a normal heart, but for older folks or cardiac or renal patients, this might be too much Na. Be aware of this.) (note K is added to fluids to replace losses.) The 4-2-1 rule. The rate of maintenance infusion is calculated in cc's per hour, and is based on the patient's weight in kg. The calculation is: 4 cc's per kg, per hour, for the first 10 kgs of weight. 2 cc's per kg, per hour, for the next 10 kg, and 1 cc per kg per hour for the rest of the weight. A short-cut based on the above is "40 plus the weight in kg's." A 70 kg man would need 40+70=110 cc's/hr of fluid, or (4 x 10) + (2 x 10) + (1 x 50) = 110 cc's/hr. Resuscitation fluids. LR or NS. Note you should not use D5 up front, because patients under stress could get quite hyperglycemic, thanks to their high stress, high cortisol, and thus, high counter-regulatory activity. They don't need the extra sugar. A common error in resuscitation is to run fluids too slow. In a very dehydrated adult, it is acceptable to give a 1-2 liter bolus over 30-120 mins, UNLESS the patient has a cardiac history, in which case a slower rate may be prudent. In children, an acceptable bolus would be 10-20 cc's/kg. Note the above fluid choices will vary depending on the particular clinical scenario. Admitting a patient: Meg’s List On call nights, you will help admit a patient, including doing your own H&P. I'm not going to go into how to do that-except to reassure you that you CAN do it, and that I strongly recommend Sapira's as a guide for enhancing basic H&P skills. I do have some pre- and post-admission tips. This may seem obvious, but when your R2 says "I've got a patient for you," make sure you get the patient's name and date of birth (DOB), and/or medical record number. Your R2 may not have all of this info, but try to get it-it may be important in locating patients who have already been sent up from the ER. If there's time, you will be able to read previous discharge summaries from the patient's computerized medical record, giving you a leg up After admission, it's a good idea to review some issues that are commonly forgotten in the rush of admission, esp. in the wee hours. The following checklist is called "Meg's List," and was created by one of my mentors, Dr. Meg Newman. She hands it out to all her teams. You should glance at Meg's list every day when you think about your patients to avoid common management errors. Note this version of Meg's List was current as of Autumn, 2000. Renal dosing. If this patient has a kidney disorder, they will likely need lower doses of common meds. Meds needing adjusted dosing for people with kidney disease include, but aren't limited to: Antibiotics: aminoglycosides, most cephalosporins, penicillins, Bactrim, vancomycin H2 blockers: cimetidine, ranitidine, famotidine NSAIDS-CAUTION! PGE2 inhibition can put underperfused kidneys into crisis Digoxin Mg in antacids Precautions. Some patients may need orders for special sets of standard precautions by nursing staff. The order for these would be "Aspiration precautions" or similar. They include: aspiration precautions - e.g. past or acute stroke with oral dysmotility seizure precautions - e.g. alcoholic with past seizure being treated for withdrawal fall - e.g. any patient with altered gait DVT -e.g. patient who will not be able to walk due to surgery, stroke, mental status IV lines. Of course, lines (and other invasive ports, like chest tubes) should be part of your standard pre-round 5-point physical exam (chest, heart, abd, wound, and lines). Check for infiltration, thromboplebitis, ascending infection. Peripheral lines must be changed every 3 days. Line type Life Femoral -considered the "dirtiest" site 3-4 days maximum Internal jugular -supposedly cleaner than fem lines, but evidence says they are equivalent 4-5 days; convention says up to 2 weeks, esp. if placed by interventional radiologists Subclavian 2 weeks PICC Line Up to 3 years! Line Type Life Femoral -considered the "dirtiest" site 3-4 days maximum Internal jugular -supposedly cleaner than fem lines, but evidence says they are equivalent 4-5 days; convention says up to 2 weeks, esp. if placed by interventional radiologists Subclavian 2 weeks PICC Line Up to 3 years!     Ordering antibiotics. When admitting a sick patient with active infection, make sure to write an order for "STAT" dose, and VERBALLY COMMUNICATE this with the floor nurse when the patient arrives on the floor. Otherwise, the dose will not happen, and in some cases this delay could be very harmful. Know that you must simultaneously fill out an "AOS" or "antibiotic order sheet" to start the meds. You should know that if you do not check off that the patient has a "documented infection" (as opposed to "empiric treatment") on this form, the pharmacy will automatically shut off the antibiotics after three days. This is a real headache for the managing team, since by then you've forgotten about this and sometimes the drugs are missed for a half day or more. Also note that I'm not saying you've actually got to have documented the infection at the time you fill out the form this way-I'm just telling you what will happen if you don't check that box. Labs. Don't bleed patients unnecessarily. Make sure all the daily labs you're getting are pertinent to the patient's care, and cut down on labs and their frequency as soon as you can. Ulcer prophylaxis. Is the patient at risk for GI stress ulcers? Are they on a PPI (proton pump inhibitor) or H2 blocker? Foley, a.k.a urinary catheter. If the cath has been in place 5-7 days, you must taper off its use before totally removing it, since the neurological signal controlling bladder function may have been disturbed by prolonged placement. Clamp the catheter for 2-4 hours with a hemostat (DON'T forget to remove the hemostat), then open it, allow urine flow. Later, remove the Foley. ALSO: Watch for UTI's from prolonged use of Foleys-many an older patient has died from urosepsis due to catheters after surviving other problems. Skin care. Look for ulcers, esp. for immobile patients, esp. around the sacrum. Commonly missed diagnoses. Stumped by your patient? Consider: CV: MI, pericarditis Pulm: PE/DVT ID: meningitis, spontaneous bacterial peritonitis (SBP), endocarditis, occult abscess (ear, sinus, oropharnyx, rectal, skin) Neuro: CNS bleed, encephalopathy Heme: DIC Endo: hypothyroidism, adrenal insufficiency GI: hepatic encephalopathy Drug interactions and side effects. If a new problem emerges during hospitalization (or indeed anytime), always consider drugs high on your list of possible causes. Blood cultures. Drawing blood cultures from established lines raises concern for contamination. Femoral lines are the dirtiest site. Pain meds. Change from I.V. to p.o. form ASAP. Titrating pain management with oral meds must be completed before patient can be discharged. And never forget: Dispo, dispo, dispo. If patient is homeless, frail eldery, has suffered serious event likely needed long-term convalescence, etc. give your social worker a head's up ASAP. The Blood Draw in 12 Steps The level of formal instruction in all procedures by UCSF School of Medicine for med students is grossly inadequate. Just as an example, drawing blood well and safely requires practice-which is not something you get with a 2-hour tutorial a few days before being thrown into the wards. This produces students who don't feel confident in their skills and who may therefore not seek out opportunities to practice. In my mind, at least for the skill of drawing blood, med students should get a half-day training in blood draws and IV placement in year 1 or 2 and have pre-assigned times in the ER to practice these skills. Currently this is an option only if you take the "Emergency Procedures" elective. That's all the more reason you should seek out opportunities to practice procedures when you get to the wards. You should always be supervised when doing procedures, unless you're doing something you've done dozens of times. Rule of thumb: After three failed attempts, get someone else to try. Below is a list of steps for drawing venous blood. I memorized these steps and would mentally run through them before every draw, so I wouldn't forget anything. Gather materials. Mentally run through the steps and identify the materials you'll need. They are: gloves, alcohol pad, the Vacutainer test tubes in which the blood will go, a butterfly or "direct" style needle, a yellow cup adapter for the needle, stickers with the patient's MRN (medical record number) and name, a tourniquet, a cotton ball or bandage for applying pressure to the site post-draw, a band-aid, and an accessible biohazard box for disposing of the needle. Make friends with the vein (and the patient!) This means identify the vein you will draw. Introduce yourself to the patient, explain what you will be doing, and ask for their assent and advice on a draw site. If the patient is a "frequent flyer" or IVDU, they will be able to advise you which vein has a high success rate. (Note: The patient is always right. In my brief clinical career, I have never seen a case where the patient was wrong, and many cases in which nurses/residents insisted on trying to draw from veins that the patient insisted wouldn't work, much to the consternation of the patient-and to the clinician, when they failed after many attempts.) For students, the easiest and least painful sites are the basilic and cephalic veins, on the medial and lateral sides, respectively, of the antecubitum (inner elbow). Palpate the various options with your fingers. Note you should do this gloveless, to enhance your sensitivity. Being able to see the vein is a good sign, if it's large and slightly springy to the touch. Some little veins that are seen but not palpable are often tricky. Veins that can't be seen but can be felt, on the other hand, often work great. In other words, feeling a vein is more helpful than seeing it. There are several tricks to help "evoke" the veins, including: rubbing and patting the potential site, to warm the vessels and make them more prominent; tying on the tourniquet (temporarily!) to help engorge the veins and make them more visible; using an adjustable lamp to light the site from the side, which will make the vessel contour more obvious. Some folks rub alcohol on the site, but since evaporation would cool the skin and should make vessels shrink up, I would think this would hinder rather than help. Put on gloves. They should be comfortable-neither too tight or loose. Double gloving can make your fingers MUCH less sensitive; I think this is overkill and will reduce the chances of a safe and successful draw. Rub the site with alcohol. My algorithm has three steps between applying alcohol and inserting the needle under the skin. There's a reason. Many phlebotomists rub alcohol on the site IMMEDIATELY before they draw blood. This is an excellent way to ensure an uncomfortable draw, since it stings like a mother. It's probably done by people who have never had their blood drawn in this fashion. LET THE ALCOHOL AIR DRY BEFORE YOU STICK. The Golden Rule of Procedures: "Do the procedure unto others as you would have the procedure done unto you." Label the tubes. This is the second most important step of the draw. Assemble the needle. I advise students to use the "butterfly" type needles, which are easier to handle and more comfortable for patients. These have a little butterfly-shaped plastic "handle" along the base of the needle, and a long thin flexible plastic tube attaching the needle to a rubber-coated "barb" that will squirt out the blood. The barb needs to have a little yellow plastic cup (like a bucket with a hole in the bottom) screwed over it. Blood tubes then get plugged into the cup. The tubes get stuck into the cup with their rubber caps going in first. As the tube top touches the bottom of the cup, the barb sticks through the hole and penetrates the tube top. Since the Vacutainer tubes have a mild vacuum (thus the name), they suck blood through the butterfly needle. This will all be more obvious in a live demo. Apply tourniquet. How to do this will be demonstrated during your orientation training. Doing it deftly requires practice. Tip: At the beginning, I was often shy about putting the tourniquet on tightly, since I didn't want to hurt the patient. While you don't want it to hurt, and should remove and re-tie it if the patient asks you to, it should be as snug as possible-patients who I thought had no good veins were actually "easy sticks" once I started applying the tourniquet tightly, which helped me find the veins in steps 1 and 7. Uncap needle, and insert. Remember, 45 degree angle, along the course of the vein. If you don't get a "flash," don't panic. Sloooooowly pull back on the needle until it's almost (but NOT QUITE) out of the skin, and re-direct the needle, again slowly, along a slightly different path. Watch for a flash at all times, and when you get it, hold the needle steady with your dominant hand and plug the tubes into the yellow cup with your other hand. Getting a flash is all about practice. Fill tubes, and invert. That is, once the tube is full, or has enough blood, pull it off the barb, invert upside down to mix the anticoagulants, and then (ideally) place upright in a rack, or at least on its side away from any edges off which it could roll. Untie tourniquet. If you've tied it correctly, a mild tug on the band should let it snap off. Apply cotton ball. Or 4-by-4 or other bandage. Note: Many phlebotomists PUSH DOWN on the needle site with the cotton ball. As with the hasty alcohol application in Step 3, this is poor form: Pulling the needle out of the arm while you are pressing on the area is going to make it hurt more. Just place the cotton ball so it lightly covers the area where the needle is inserted. Once the needle is out, you can press on the cotton, and ask the patient to apply pressure to the cotton for you so you can free your hands for needle disposal. Pull out needle and place in biohazard bucket. THIS IS THE MOST IMPORTANT STEP OF THE DRAW. Nothing else in the draw really matters, other than to dispose of the needle safely. Even if you don't get a flash, don't fill a single tube, whatever--it doesn't matter, as long as you avoid sticking yourself or anyone else with the needle. If you fail to draw the blood, someone else will get it. If you have a needlestick, however, it can ruin your day--or your life. By the way, the UCSF Needlestick Hotline is 415-353-2611. Once the needle leaves the skin, your eyes should never completely come off of the needle until it's safely in the bucket. If other personnel are in the room with you, and the bucket is far enough away that you have to step over to it, you should announce, loudly and clearly, "I have a sharps," and keep one hand cupped along the side (NOT in front of) the needle tip. Then quickly glance to see if the path to the bucket is clear, and dispose of the needle. Apply band-aid. Thank the patient for their cooperation. Congratulations... Of course, you must still fill out the lab requisition form, completely and properly. The date and time of the draw, your name, your resident's name and pager number (key, in case there's a problem), and the requested tests should be submitted with the tubes. Ask for help if you're unsure. [20] Finger stick blood sugar. [Back] [21] House officer, i.e. intern or resident. [Back] [22] Gearhart JG et al. Efficacy of sliding-scale insulin therapy: a comparison with prospective regimens. Fam Pract Res J 1994;14:313-22. [Back] [23] Queale WS et al. Glycemic control and sliding scale insulin use in medical inpatients with diabetes mellitus. Arch Intern Med 1997;157:545-52. [Back] [24] Not "alkalotic/acidotic," until you examine what's producing the pH. [Back]