The anesthesiologist and surgical site infection!

A 65 year old male for emergent hemicolectomy secondary to massive bleeding. Pt has remote history of testicular cancer (age 30), now with HTN, DM and obesity (BMI 32). Current BP is 100/70, HR is 85, AF, RR 27. Plan is GETA w/ RSI and Cricoid Pressure.The anesthesiologist often does not consider surgical site infection as a main theme or criteria when deciding what medications to give or in developing the anesthetic plan. In many cases, issues of surgical site infection (SSI) are simply to give preop antibiotics and go no further. However, there are special cases as the one above, where your patient has an above average potential for developing a surgical site infection and the anesthesiologists role is directly relevant to this outcome. Although, maintaining normothermia, adminstering effective prophylactic antimicrobials prior to incision, and maintaining blood flow to tissues are obvious and well known to be critical in reducing the risk of surgical site infections, there are other methods that are important, particularly in higher risk patients that are less known and understand by many anesthesia providers who have a great amount of experience.The first step in playing an active role in reducing the probability of a SSI post op is estimating your patients risks which implies understanding those factors that increase risk.the following is a general list of risk factors but does not quantify how much each factor increases risk:

hypovolemia

diabetes

obesity

malnourishment

blood trasnfusion

inadequate pain control

antibiotic prophylaxis

prolonged surgical time

host immune system

air in OR (ultraclean, negative pressure)

hypothermia

low SC Oxygen tension

This article will focus on the few modifiable risk factors that exist with an emphasis on those that are predominantly under the control of the anesthesiologist. These will include:


pain control-brief

antibiotic prophylaxis-brief

patient temp-brief

Sub Q oxygen tension- emphasis

Pain Control: at this point I will limit my comments to a simple better pain control is better. The 'why' will be touched upon under the heading of Sub Q oxygen tension (improved w/ pain control).

Antibiotic prophylaxis: appropriate antibiotics should be given within one hour of surgical incision. Giving antibiotics after incision is has questionable benefit.

Patient temp: normothermia is critical. This is often very difficult in a scenario as described above because patients come to the OR emergently and time is spent placing lines in a pt who is not being warmed in a cold OR. Often by the time the drapes go up and surgery commences the patient has become hypothermic. Evidence for the importance of normothermia comes from a RCT published in the NEJM demonstrating an absolute risk reduction (ARR) of 13% by maintaining normothermia (37C vs. 34.5C)(1).
Sub Q oxygen tension: Most anesthesiologists do not consider this data when performing anesthesia. Indeed, it is usually not important information for the vast majority of anesthetics. However, as arterial lines are appropriate to place in some patients, the information gained from sub q oxygen tension can be quite important. Unfortunately there is no clinical method available to monitor this parameter. Nevertheless, it is important to understand what is normal, what is abnormal, and how this will affect your patient. In a normal volunteer breathing room air, it sits around 65 mmHg. Surgical patients when measured average about 49 mmHg with large variability. Post surgical patients breathing oxygen through NC (~0.4 to 0.6 FiO2) average around 69 mmHg, whereas a normal volunteer given the same amount of oxygen will increase their TsqO2 to ~130 mmHg (2). This is important to understand because it turns out that tissue healing and bacteria killing are dependent on an oxygen tension in the tissues and vary in direct proportion to the tissue oxygen tension (PsqO2). In fact it has been experimentally verified that oxidative killing is oxygen dependent from 0 to 150 mmHg PsqO2 (3). Furthermore, when bacteria are introduced and phagocytes begin utilizing NADPH to reduce O2, this results in a dramatic reduction in PsqO2, that is from the normal of ~60 mmHg down to 0-10 mmHg. Bacterial contamination also alters PsqO2 by altering perfusion of tissues independently of oxygen utilization. In rats where lesions have been experimentally created and then inoculated it has was demonstrated that low FiO2 compared to high FiO2 was an independent variable in determining infection size and was no less important than giving prophylactic antibiotics. In other words, utilizing a FiO2 after bacterial inoculation was just as effect maintaining normal (0.21) FiO2 but giving antibiotics (4,5).
So who do we give oxygen to and in what dose? The first step that should be taken in any patient going to the OR where a surgical incision will be made is to make a valid determination of the risk of SSI to the patient. Although the above risk factors as listed are pertinent, considering them provides no quantitative probability of a particular patient developing a SSI post operatively. A crude, but simple tool that has been used in research is the SENIC scoring system. See below:
1 pt: intraperitoneal abdominal surgery
1 pt: duration of surgery >2 hrs.
1 pt: 3 or more co morbidities (i.e. hypertension, pulmonary disease, diabetes)
1 pt: contaminated wound/surgery- (i.e. abscess, colon surgery non prepped)
Predicted infections rates are as follows depending on score:
0=1% 1pt=3.6% risk 2pts.=9% 3pts.=17% 4pts.=27%
However, in 1997 Hopf and colleagues published a paper in the Archives of Surgery demonstrating that this scoring system was not nearly as effective at predicting the risk for SSI as was measuring the PsqO2 on patients(6). Among their findings were the following:
· If baseline PsqO2 was >70 mmHg, no SSI occurred.
· If baseline PsqO2 was 40 mmHg or less, SSI rate was 40%
· Rate of infection varied inversely with both baseline PsqO2 and with post operative maximal PsqO2.
There are three main factors which determine PsqO2, two of which are under the control of the anesthesiologist. These are vascular anatomy at the site of inoculation (surgical site), vasomotor control (vasoconstricted vs. vasodilated), and PaO2 which varies directly w/ PsqO2 assuming normal blood flow to tissues.
The first factor is beyond our control, however, vasomotor control and the PaO2 can be manipulated by us to a much greater degree.
Controlling PaO2: Two large RCTs in colon surgery patients both demonstrated that indeed, by providing a FiO2 (80% vs. 30%), and therefore increasing PsqO2, indeed decreases the rate of SSI in colon surgical patients although the decrease was modest (7,8). Patients in both of these large studies had significantly lower SSI rates when given 80% FiO2 both during surgery and afterwards for 2 or 6 hours depending on the study.
Vasomotor state: Okca et al(9) demonstrated that in patients undergoing surgery w/ FiO2 of 0.4 received an average increase of ~30 mmHg (from 63 to 89 mmHg) if the etCO2 was maintained at 45 mmHg vs. 30 mmHg. Note that is quite common for the etCO2 to be maintained at ~30 mmHg in patients who are paralyzed and mechanically ventilated. Patients already receiving FiO2 of 0.8 also benefit however(10), going from PsqO2 of 84 mmHg to 116 mmHg and more importantly, an intestinal oxygen tension went from 53 mmHg to 107 mmHg.
Hypovolemia is detrimental to PsqO2. In fact, if you give provide a higher FiO2 to a patient with goal of raising PsqO2 and thus decreasing chance of SSI, but do not adequately resuscitate your patient, it has been shown that despite your good intentions, the PsqO2 will not increase. It’s only after adequate volume status has been restored that raising FiO2 will prove beneficial (11). Others have shown that indeed an aggressive fluid regimen can augment PsqO2 in abdominal surgical patients (12), but a follow up study on this same cohort of patients found that this increase in PsqO2 did not result in a decrease in the rate of SSI (13). Despite these findings, hypovolemia must be avoided and fluid therapy should be adequate to ensure this. In the above study looking at an aggressive fluid regimen vs. a standard protocol of 8 to 10 mL/kg/hr a balanced crystalloid solution was used. However, Lang et al (14). showed that LR in large volumes (11L) caused a 59% decrease in the PsqO2, however, giving a HES in addition to the LR at lower volumens (3 L each) PsqO2 was increased by 23%. The take home message from these studies is avoid hypovolemia, but caution she be used when this requires large volumes of crystalloid (i.e. >8 or 9 L) and consideration should be given to other fluid types.
Inadequate pain control has been shown to have a significant effect on decreasing the PsqO2 by around ~24 mmHg (15). Morphine and its derivatives effective at reducing pain, however, they also can cause respiratory depression and lead to decreased PaO2 which reduces PsqO2. Furthermore, meta analyses have shown that PCA with narcotic pain medications are inferior to PCEA after many different kinds of surgery with regard to pain scores (16, 17). Therefore, whenever appropriate regional anesthesia should be considered as a method not only for pain control but also as an intervention with the goal of decreasing SSI risk. In addition to the improvement provided by improved pain control to the PsqO2 of nearly 24 mmHg as shown, the vasodiliation that results from the sympathectomy of regional anesthesia induces a further increase in PsqO2 of from 9 mmHg to 31 mmHg depending on the method and study (18, 19, 20, & 21).
Obesity is associated with a greater risk of post op SSI. Studies looking at the PsqO2 in obese patients (BMI >30), showed that to no one’s surprise, a much greater FiO2 was required to maintain a certain PaO2 (22, 23). However, more surprising was the studies findings that once obese and non obese patients achieved an equal PaO2, PsqO2 were still much lower in the obese patient by approximately 20 to 30 mmHg depending on the PaO2 that was used as the starting measurement. Both studies indicated that obese patients require FiO2 of >0.95) in order to maintain PsqO2 around ~47 mmHg, which is still low enough to be associated with risk. Furthermore, simply raising FiO2 in this population results in an underwhelming increase of PsqO2 of 13 mmHg, which is half the benefit their non obese counterparts experience. Adding PEEP may be necessary if tolerated in order to increase PaO2 to >300 mmHg, the maximum tested in these studies.
Although PsqO2 is an excellent predictor of risk of SSI, and two RCT have demonstrated benefit to hyperoxia, the practice of delivering high FiO2 is not standard of care. Many cite a study showing that hyperoxia was associated with worse outcome in a study of general surgical patients (24). This study however suffered from methodological flaws: the experimental group (hyperoxia) had a greater number of obese patients, were more likely to be intubated at the end of surgery (likely sicker) and underwent longer and more complicated surgery. Simply put, the groups were not equal. Nevertheless, performing a metanalysis of the three studies shows that hyperoxia is beneficial. Another common concern among clinicians is that of absorption atelectasis resulting in post operative pulmonary complications. Although, this is a known feature of high FiO2, a study looking specifically at this problem in a clinical setting demonstrated that this concern is likely unfounded (25).
In conclusion, it is my current practice to utilize every maneuver possible to maximize PaO2 in at risk patients who present to the OR. I generally estimate risk utilizing a combination of factors including the SCENIC score (although I do not necessarily chart of format pt value for each patient), surgical setting (emergent/elective), type of operation (bowell vs. highly vascular site) surgical duration etc. In most cases FiO2 for short duration is harmless, and therefore that is my default. I typically avoid Nitrous (not necessarily because N2O has been shown to be a risk factor for SSI itself [26], but because it requires FiO2< 1.0. I assiduously avoid hypovolemia in at risk cases (large abdominal case) and use hespan is necessary. I am careful to maintain etCO2 of 45 mmHg in at risk cases per the studies mentioned above. I always offer epidural anesthesia and discuss the risks and benefits with those patients what would benefit. SSI can be devastating for patients, results in greater mortality, and cost to our healthcare system. The anesthesiologist is plays a critical role in this outcome since it has been demonstrated that the critical or decisive period is within two hours of inoculation, for antibiotics and up to 6 hours with regards to oxygen therapy.
I want to conclude with a note on blood transfusion. It is well known that blood transfusion results in immunosuppression and therefore should be avoided in those patients who are at risk for SSI. Some may be tempted to transfuse blood to someone who with a decreased Hbg in order to improved oxygen delivery (DO2) to the tissues so that neutrophils can have substrate to perform their bactericidal duties. However, while DO2 is certainly a function of Hgb, this is only true in vessel rich tissues, but is not at all the case in wounds where intercapillary distances are large. At these sites, oxygen delivery is more a function of adequate perfusion and oxygen tension (i.e. PaO2 and PsqO2). Therefore, transfusions of blood should only be considered if DO2 to critical organs is deemed inadequate.
1. Kurz A et al. Perioperative Normothermia to reduce the incidence of surgical wound infection and shorten hospitalization. NEJM Vol 334, No. 19, 1996
2. Hopk HW, Hunt TK, West JM et al. Wound tissue oxygen tension predicts the risk of wound infection in surgical patients. Arch Surg. 1997;132:997-1005
3. Allen DB, Maguire JJ et al. Wound hypoxia and acidosis limit neutrophil bacterial killing mechanisms. Arch Surg. 1997 Sep;132(9):991-6.
4. Knighton DR, Halliday B, Hunt TK. Oxygen as an antibiotic: the effect of inspired oxygen on infection. Arch Surg. 1984; 119:199-204.
5. Knighton DR, Halliday B, Hunt TK. Oxygen as an antibiotic: a comparison of the effects of inspired oxygen concentration and antibiotic administration on in vivo bacterial clearance. Arch Surg. 1986. 121:191-195.
6. Hopk HW, Hunt TK, West JM et al. Wound tissue oxygen tension predicts the risk of wound infection in surgical patients. Arch Surg. 1997;132:997-1005.
7. Greif R, Akca O, et al. Supplemental Perioperative Oxygen to Reduce the Incidence of Surgical-Wound Infection. N Engl J Med. 2000 Jan 20;342(3):161-7.
8. Dellinger PE. Increasing Inspired Oxygen to Decrease Surgical Site Infection. JAMA. V. 294, N. 16 Oct 2005.
9. Okca O, et al. Effect of intraoperative End tidal CO2 on oxygen tension. Anesthesia, V. 58, 2003.
10. Edith F, Friedrich H et al. Mild Hypercapnia Increases SQ and Colonic oxygen tension in pts given 80% FiO2 during abdominal surgery. Anesthesiology. 2006 V 104: 944.
11. Jonsson K et al. Assessment of perfusion in postoperative patients using tissue oxygen measurements. Br J Surg. 1987 Apr;74(4):263-7.
12. Arkilic C. Supplemental perioperative fluid administration increases tissue oxygen pressure. Surgery Jan 2003
13. Kabon B, Akca O, Taguchi A. Supplemental IV Crystalloid administration Does Not Reduce the Risk of Surgical Wound Infection. Anes Analg. 2005;101:1546
14. Lang K, et al. Colloids vs. Crystalloids and Tissue Oxygen Tension in Patients undergoing Major Abdominal Surgery. Anesth Analg 2001; 93:405
15. Akca O et al. Post-operative Pain and SQ oxygen tension. Lancet 1999; vol 354: 42
16. Block BM, Liu SS, Rowlingson AJ, Cowan AR, Cowan JA Jr, Wu CL. JAMA. 2003 Nov 12;290(18):2455-63.
17. Wu CL, Cohen SR, Richman JM, Rowlingson AJ, Courpas GE, Cheung K, Lin EE, Liu SS.Anesthesiology. 2005 Nov;103(5):1079-88; quiz 1109-10.
18. Treschan TA et al. Effects of Epidural and General Anesthsia on Tissue Oxygenation. Anesth Analg 2003; 96: 1553
19. Buggy DJ. Paraveterbral anesthesia increases post op flap tissue oxygen tension….compared to IV opioids analgesia. Anesthesiology 2004; 100;375
20. Buggy DJ, Dougherty DL. Postoperative Wound oxygenation with epidural or intravenous Analgesia. Anesthesiology 2002; 97:952.
21. Kabon B et al. Thoracic Epidural Anesthesia Increases Tissue Oxygenation During Major Abdominal Surgery. Anesth Analg 2003; 97:1812 .
22. Fleischman E. et al. Tissue oxygenation in obese and non-obese patients during laparoscopy. Obes Surg. 2005 Jun-Jul;15(6):813-9
23. Kabon B, Nagele A, Reddy D, Eagon C, Fleshman JW, Sessler DI, Kurz A. Obesity decreases perioperative tissue oxygenation Anesthesiology. 2004 Feb;100(2):274-80.
24. Pryor KO, Fahe TJ et al. Surgical Site infectin and the Routine Use of Perioperative Hyperoxia in a General Surgical Population: RCT. JAMA. Vol 291, No.1 Jan 2004.
25. Akca O, Podolsky A, Eisen huber E et al. Comparable postoperative pulmonary atelectasis in patient given 30% or 80% oxygen during and 2 hours after colon resection. Anesthesiology 1999;91:991-8.
26. Fleischmann E. Et al. Lancet 2005;366:1101-1107