M. Meisner K. Tschaikowsky A. Hutzler C. Schick J. Schiittler
Postoperative plasma concentrations of procalcitonin after different types of surgery
Received: 20 October 1997 Accepted: 25 March 1998
M. Meisner ([~) University of Jena, Department of Anaesthesiology, Bachstrasse 18, D-07743 Jena, Germany Tel.: + 49-3641-933041 Fax: + 49-3641-93 3256
K. Tschaikowsky. A. Hutzler • J. Schiittler University of Erlangen-Nuremberg, Department of Anaesthesiology, Krankenhausstrasse 12, D-91054 Erlangen, Germany
C. Schick University of Erlangen-Nuremberg, Department of Surgery, Maximiliansplatz 2, D-91054 Erlangen, Germany
Abstract Objective: Procalcitonin (PCT) and C-reactive protein (CRP) plasma concentrations were measured after different types of surgery to analyze a possible post- operative induction of procalcitonin (PCT), which might interfere with the diagnosis of bacterial infection or sepsis by PCT. Design: PCT and CRP plasma levels as well as clinical symptoms of in- fection were prospectively regis- tered preoperatively and 5 days postoperatively. Setting." University hospital, in-pa- tient postoperative care. Patients: Hundred thirty patients were followed up; 117 patients with a normal postoperative course were statistically analyzed. Interventions: None. Measurements and results: PCT con- centrations were moderately in-
creased above the normal range in 32 % of patients after minor and aseptic surgery, in 59 % after cardiac and thoracic surgery, and in 95 % of patients after surgery of the intes- tine. In patients with an abnormal postoperative course, PCT was in- creased in 12 of 13 patients. CRP was increased in almost all patients. Conclusions: Postoperative induc- tion of PCT largely depends on the type of surgery. Intestinal surgery and major operations more often increase PCT, whereas it is normal in the majority of patients after mi- nor and primarily aseptic surgery. PCT can thus be used postopera- tively for diagnostic means only when the range of PCT concentra- tions during the normal course of a certain type of surgery is considered and concentrations are followed up.
Introduction
The infection-induced protein of procalcitonin (PCT) is a diagnostic parameter for differential diagnosis of bac- terial and non-bacterial infections of patients at risk of infection. After surgery, especially after major opera- tions and in immunocompromised patients, the risk of infection is increased. PCT measurement for diagnostic purposes would only be helpful in patients at risk when the surgical trauma does not induce considerable amounts of PCT. Experimental data [1, 2] and clinical observations [3-8] indicate that the main stimulus of PCT is bacterial endotoxins. PCT can be induced by oth- er stimuli to a minor extent, but is thought not to be in-
duced by a surgical trauma per se. At present, however, the mechanisms of induction are not fully understood.
Occasionally~ increased PCT concentrations were ob- served after surgery. Elevated PCT concentrations were reported after esophagectomy especially [9]. Bacterial endotoxins originating either from bacterial transloca- tion or due to a transient bacterial contamination during the operation, e.g. by preparation of anastomoses, might induce PCT. Furthermore, other mechanisms of induction, e. g. by cytokines during the course of wound healing, may also contribute to postoperatively in- creased PCT. In multiorgan dysfunction syndrome (MODS) and after thermal injury, for example, in- creased PCT can be observed in some cases, even if no
681
bacterial infection is evident [7-12]. Since there are only few data available on postoperative PCT plasma con- centrations, we prospectively analyzed PCT in patients with a normal postoperative course without infection, inflammation or other complications. The following data provide information about normal PCT plasma concentrations observed after different types of surgery, which is important for the interpretat ion of postopera- tively increased PCT concentrations.
Methods
One hundred thirty patients undergoing surgery were enrolled into the study and assigned to five different categories of surgery ac- cording to the type of operation (Table 1), and studied prospec- tively. The following data were registered preoperatively (POD 0) and postoperatively on days 1-5 (POD 1-5): PCT, CRR white blood counts, body temperature (oral) and clinical signs of infec- tion or inflammation including microbiological findings. In addi- tion, the ACCP/SCCM definition criteria for SIRS and sepsis were evaluated in all patients. Patients who matched SIRS or sep- sis criteria, but also those with any clinical signs of infection, in- creased body temperature above 38.0 °C after POD 2, extubation after POD 1, catecholamine therapy, cardiac insufficiency, pulmo- nary infiltrates, local wound infection or reoperation, were exclud- ed from the statistical evaluation, but are shown as case reports (n = 13, Table 2). Hundred seventeen of the observed patients had a normal postoperative course and were evaluated for the study. The study was approved by the institutional Ethics Committee. In- formed consent was obtained from all patients.
In each group, the mean, median and 25 th, 75 th and 90 th per- centiles of the investigated parameters were calculated for each postoperative day. In addition, maximal plasma concentrations of PCT and CRP during the entire postoperative period were deter- mined. Statistical difference among the groups was calculated by the Mann-Whitney test (MWU test). For statistical analysis, the program Statistica 4.5 for windows was used.
Results
In a l l groups plasma concentrations of PCT above the normal range (_< 0.5 ~tg 1 i) were observed (Table 3). However, the incidence and the median values of in- creased PCT plasma concentrations showed a wide vari- ation depending on the type and extent of surgery. After minor and aseptic operations, e.g. peripheral bone or vascular surgery, increased PCT concentrations were de- tected in 32 % of the patients, but only seldom (< 8 %) were they above 1 ~tg 1-1. Similar results were observed during minor abdominal surgery (p = 0.37, M W U test). After cardiac and thoracic surgery, PCT increases above 2 ~tg 1-1 occurred only in 8 % of the patients, and re- mained below 0.5 ~tg 1-1 in 41% of patients. PCT con- centrations after cardiac and thoracic surgery were thus significantly higher than those after minor primarily aseptic surgery (p < 0.001), but less than those after sur- gery of the intestine (p < 0.002).
Table 1 This table indicates the categories of surgery the patients were assigned to according to the type of operation (n number of patients)
1. Minor, primarily aseptic surgery (n = 37), including: - Total hip replacement (n = 14) - Peripheral vascular surgery (n = 15, including carotic surgery
(8 patients), femoro-popliteal bypass (5 patients), crossecto- my (2 patients)
- General surgery (n = 8, including thyroidectomy (4 patients) and hernia surgery (4 patients)
2. Minor abdominal surgery (11 patients with cholecystectomy)
3. Abdominal surgery of the intestine (n = 22), including: - Resection of the colon (7 patients) - Resection of the sigma (3 patients) and rectum (4 patients) - Gastrectomy (8 patients)
4. Major abdominal or thoracic surgery affecting the mediastinum or retroperitoneum (n = 16), including: - Esophagectomy (8 patients) and Whipple's operation (2 pati-
ents) - Major vascular surgery (n = 6, including Y-prosthesis (3 pati-
ents), abdominal or thoracic aortic aneurysm (3 patients each)
5. Cardiac and thoracic surgery (n = 44), including: - Coronary artery bypass grafting or valve replacement (n = 32) - Thoracic surgery without EEC (n = 12, including resection of
the lung (5 patients) and diagnostic thoracotomy (7 patients)
Interestingly, after abdominal surgery with anasto- moses of the intestine, 95 % of patients had increased PCT. In this group, PCT plasma concentrations were above 2 ~tg 1- 1 in 25 % of the patients. High concentra- tions of PCT were also occasionally observed after ma- jor abdominal or abdomino-thoracic surgery affecting the mediastinum or retroperi toneum. The PCT concen- trations of both groups were significantly higher than those after minor abdominal or primarily aseptic sur- gery (p < 0.0001). Following esophagectomy or Whip- ple's operation, two of seven patients showed increased PCT plasma concentrations. After thoracic or abdomi- nal surgery of the aorta (n = 5), the median of the PCT plasma concentration was 1.65 btg 1-t with a maximum of 5.76 ~tg 1-1. The concentrations of this group were not significantly different from those after cardiac and thoracic surgery (p = 0.963).
In contrast, CRP plasma concentrations were post- operatively elevated in almost all patients, regardless of the type of operat ion (Table 3).
Patients with an abnormal postoperative course were excluded from the statistical evaluation, From 130 pati- ents enrolled in the study (Table 1), 13 patients had an abnormal postoperative course according to the above definition, but only one patient had a microbiologically proven bacterial infection (Table 2). Twelve of these pa- tients had increased PCT. The median of PCT was 6.48 btg 1-*, the 90% percentile was 21,5 ~tg 1-1. Since this group was quite heterogenous, the data are shown
682
Table 2 This table presents the specification of the patients with an abnormal postoperative course, e. g. delayed recovery or clinical or laboratory signs of infection or inflammation. PCT (CRP) maximal plasma con- centrations (PCT-max., ~g 1 1 CRR mg 1 1) during the first postoperative days, the day of maximum values, the underly- ing diagnosis and the suspected cause for the abnormal postop- erative course are indicated (PCT procalcitonin, CRP C-re- active protein, ICU intensive care unit, POD postoperative day)
Prolonged ventilatory support (POD 1-3), temperature 38 °C (POD 1-3), catecholamines (POD 1-2)
Respiratory insufficiency, pneumothorax (POD 2), catecholamines (POD 1 and 3-4)
Reoperation because of postopera- tive bleeding (POD 1)
Reoperation because of pericardial effusion (POD 3)
Pulmonary infiltrates, temperature 39 °C (POD 3)
Respiratory insufficiency temperature 38 °C (POD 5)
Prolonged venfilatory support (POD 1-5), temperature 38 °C (POD 3-5), catecholamines (POD 1-4)
Positive blood cultures
Respiratory insufficiency, temperature 38.1 °C (POD 4)
Respiratory insufficiency, temperature > 38.4°C (POD 3-5), catecholamines (POD 2-4)
Prolonged ventilatory support (POD 1-5), temperature > 39 °C (POD 2-4), pulmonary infiltrates (POD 2), catecholamines (POD > 3)
as case reports only in Table 2. Obviously, the incidence and the levels of increased PCT plasma concentrations were greatly enhanced in these patients as compared to patients with a normal postoperative course (p < 0.0001, MWU test). In contrast, CRP plasma con- centrations in these patients were similar to the CRP concentrations of patients with a normal postoperative course (p = 0.098). Maximum postoperative plasma concentrations of PCT were mostly observed on day 1, but could also be found on days 2-5, independent of the postoperative course, whereas CRP maximum levels were not observed before day 2 (Table 4).
Discussion
We prospectively analyzed postoperative PCT concen- trations in 130 patients with different types of surgery. The incidence and the levels of postoperatively in- creased PCT plasma concentrations in patients with a normal postoperative course without infection or in- flammation depended on the type of operation. Where- as minor, aseptic surgery, including abdominal surgery, generally did not induce PCT, this occasionally in- creased to relatively high plasma concentrations after major surgery, especially after abdominal and abdomi- no-thoracic surgery.
The induction mechanisms of postoperatively in- creased PCT in patients with a normal postoperative course are not completely known. PCT may be induced
683
Table 3 Maximal procalcitonin (PCT, btg 1-1) and C-reactive pro- tein (CRR mg 1 1) plasma concentrations during postoperative days 0-5 in patients with a normal postoperative course. The pati- ents did not show signs of infection or inflammation preoperatively
or postoperatively during days 1-5 after surgery. The median, 25/75/90 percentiles and the maximal concentrations of PCT (CRP) observed in the respective group are indicated
- Thoracic surgery (resection of the lung) 12 0.31 0.58 1.61 1.64 4.96 PCT (101) (133) (159) (228) (250) (CRP)
Table 4 The day of occurrence of maximum PCT and CRP plasma concentrations during the first 5 postoperative days in 117 patients with a normal, and 13 patients with an abnormal, postoperative course are indicated (percent of patients). CRP plasma concentra- tions are not increased before POD 1 and rapidly rise on POD 2, whereas the maximum of PCT concentrations occurs most often on POD 1, but can also be observed later (POD 0 preoperative value, POD 1-5 postoperative day 1-5, PCT procalcitonin, CRP C-reactive protein)
% of patients with normal postoperative course
% of patients with abnormal postoperative course
PCT CRP PCT CRP
POD 0 0% 2% 0% 0% POD 1 43% 5% 46% 0% POD 2 24 % 69 % 23 % 46 % POD 3 14 % 18 % 7 % 23 % POD4 17% 4% 15% 31% POD 5 2% 2% 7% 0%
by a t ransient bacter ia l con tamina t ion during the opera- t ion or p repara t ion of intestinal anas tomoses , or by lib- e ra t ion and t rans locat ion of bacter ia l endotoxins during per iods of malper fus ion of the intestine, e.g. by high doses of ca techolamines or by in t raopera t ive manipula- tion. As descr ibed by E n g e l m a n n et al. [10], P C T did not corre la te with p lasma endo tox in levels when endot- oxin concent ra t ions were low. However , P C T plasma concent ra t ions were always high, when endo tox in levels exceeded 10 ng 1-1. Af t e r cardiac surgery with extracor- pora l circulat ion (ECC) , only mode ra t e ly increased p lasma endotox in levels were r epor t ed for several days after the opera t ion [11]. E n d o t o x i n m a y thus not be the only inducer of P C T fol lowing ECC. Likewise, the find- ing of e levated P C T in genera l surgery after pr imari ly sterile opera t ions or fol lowing non-abdomina l surgery suggests that media tors o the r than endotox in m a y con- t r ibute to the pos topera t ive induct ion of PCT.
Pat ients with an abnorma l pos topera t ive course more of ten had increased P C T (92 % of pat ients) than
684
patients with a normal postoperat ive course. These pati- ents had various disorders, e.g. pulmonary infiltrates, delayed recovery, reopera t ion or a positive blood cul- ture. In these patients high PCT levels were observed, which rapidly declined according to the half-life t ime [7] when no further inf lammation was present, In pati- ents with a normal pos topera t ive course, a slight in- crease of PCT was not l imited to the first pos topera t ive day. It could also be observed on postoperat ive days 2 and 3, suggesting that it is related to stimuli other than the surgical t rauma itself, such as wound healing pro- cesses, for example. Our findings of postopera t ive PCT concentrat ions are in accordance with observations re- por ted by other authors: M am i t z et al. described in- creased PCT levels after esophagec tomy [9]. PCT was found within the normal range in 97 % of patients un- dergoing cholecystectomy and other abdominal surgery [7].
In summary, postoperat ively increased PCT can be observed in some patients with a normal postoperat ive course, even if there is no obvious bacterial infection or other inf lammatory process. The incidence and the ab- solute level of increased PCT plasma concentrat ions strongly depend on the type and extent of operation.
These data on the different "normal" ranges of plasma PCT according to the operat ion pe r fo rmed have to be taken into account when PCT levels are evaluated post- operatively. PCT concentrations above 1 ~tg 1- 1 follow- ing minor or aseptic surgery or above 2 vg 1 1 after car- diac surgery were seldom observed, and patients with these levels should be moni tored carefully. After major abdominal surgery even high concentrations may occur without any signs of infection. However , PCT concen- trations exceeding 10 ~g 1-1 are very unusual in patients with an uncomplicated postoperat ive course. These pa- tients should be carefully screened for infections and complications, e.g. insufficiency of the anastomoses.
Finally, our findings suggest that PCT can be used as a diagnostic pa rame te r to indicate and to moni tor infec- tious complications in the postopera t ive period only when the particular range of concentrations after a cer- tain type of surgery is considered and concentrations are followed up, To evaluate the sensitivity and specific- ity of postoperat ive PCT concentrations for the early di- agnosis of infection or bacterially induced systemic in- f lammation, further studies according to each type of surgery are recommended.
References
1. Dandona P, Nix D, Wilson ME et al. (1994) Procalcitonin increase after en- dotoxin injection in normal subjects. J Clin Endocrinol Metab 79:1605-1608
2. Petitjean S, Mackensen A, Engelhardt R, Bohoun C (1994) Induction de la procalcitonine circulante apr~s admin- istration intraveneuse d'endotoxine chez l'homme. Act Pharm Biol Clin 13: 265-268
3. A1-Nawas B, Krammer I, Shah PM (1996) Procalcitonin in diagnosis of severe infections. Eur J Med Res 1: 331-333
4. Assicot M, Gendrel D, Carsin H, Ray- mond J, Guilbaud J, Bohoun C (1993) High serum procalcitonin concentra- tions in patients with sepsis and infec- tion. Lancet 341:515-518
6. Meisner M, Tschaikowsky K, Beier W, Schttttler J (1996) Procalcitonin (PCT) - ein neuer Parameter zur Dia- gnose und Verlaufskontrolle von bak- teriellen Entziindungen und Sepsis. Anaesthesiol Intensivmed 10 (37): 529-539
7. Meisner M (1996) PCT, Procalcitonin - a new, innovative infection para- meter. B.R.A.H.M.S.-Diagnostica, Berlin
8. Monneret G, Labaune JM, Isaac C, Bienvenu F, Putet G, Bienvenu J (1997) Procalcitonin and C-reactive protein levels in neonatal infections. Acta Paediatr 86:209-212
9. Marnitz R, Gramm HJ, Zimmermann J (1997) Elaboration of Mediators of in- flammatory response after major sur- gery. Shock 7 (Suppl): 124
10. Engelmann L, Gundelach K, Pilz U, Werner M (1997) Procalcitonin (PCT) and its relationship to endotoxin (ETX) in sepsis. Intensive Care Med (Suppl 1): 680
11. Berger D, Bolke E, Huegel H, Seidel- mann M, Hannekum A, Berger H (1995) New aspects concerning the reg- ulation of the post-operative acute phase reaction during cardiac surgery. Clin Chim Acta 239:121-130
12. Carsin H, Assicot M, Feger K Olivier R, Pennacino I, Le Bever H, Ainaud R Bo- huon C (1997) Evolution and signifi- cance of circulating procalcitonin levels compared with IL-6, TNFc~ and endot- oxin levels early after thermal injury. Burns 23:218-224
