Objective:

The primary aim of this study was to determine the cardiovascular safety of topical racemic epinephrine pellets by measuring heart rate, systolic blood pressure, diastolic blood pressure, and mean arterial pressure in children receiving dental care under general anesthesia. The secondary aim was to assess clinical efficacy by measuring time to reach adequate hemostasis.

Methods:

For this pilot study utilizing a split-mouth randomized design, 13 patients requiring prefabricated zirconia crowns on both primary maxillary first molars were recruited. Patients received continuous infusions of propofol and remifentanil with 50–70% inhaled nitrous oxide and oxygen. After randomization and tooth preparation, either saline pellets (control) or racemic epinephrine pellets (experimental) were applied directly to gingival tissue. Vital signs were recorded for 5 minutes. The procedure was repeated on the contralateral side using the alternative (control or experimental) treatment.

Results:

Topical racemic epinephrine compared to saline produced a significantly larger decrease in mean diastolic blood pressure (−11.1% vs −3.9%; P < .01) and mean arterial pressure (−8.1% vs −2.1%; P < .01), although all noted decreases in cardiovascular variables were clinically insignificant. All experimental treatment teeth achieved adequate hemostasis after 2.2 minutes. Only 5 of the 13 control treatment teeth achieved adequate hemostasis during the 5-minute observation period (1.6 vs 4.2 minutes; P = .01).

Conclusion:

Overall, we conclude that use of topical racemic epinephrine pellets did not result in adverse cardiovascular effects and hemostasis was reached more quickly and predictably compared to saline pellets.

Dental caries is the most common chronic disease of childhood, and it frequently necessitates extensive restorative dental treatment.1  Prefabricated crowns are the treatment of choice for children with rampant caries involving large or multiple surface lesions or developmental defects.2  Although there are many full-coverage crown options for restoring primary dentition, prefabricated zirconium dioxide (zirconia) products are popular because of their high strength, positive gingival response, and excellent esthetics.36 

Teeth receiving prefabricated zirconia crowns require circumferential subgingival preparation, leading to tissue irritation and bleeding. If hemostasis is inadequate during cementation, blood contamination will affect the integrity of the tooth-cement-crown interface. Blood incorporated into the cement may also cause visible discoloration because of the translucent nature of zirconia, resulting in poor esthetics. Preoperative or intraoperative administration of nonsteroidal anti-inflammatory drugs (eg, ketorolac) for analgesia can compound this problem because of impaired thromboxane A2 production and the resulting potential to increase surgical bleeding.7,8 

Hemostasis can be achieved by allowing the tissues to clot naturally while applying direct pressure with gauze. However, in clinical settings it is not always possible or desirable to wait extended periods of time for hemostasis. Thus, clinicians have relied upon topical or injected vasoconstrictors to expedite this process. In medicine and dentistry, dilute formulations of injectable epinephrine are commonly used for local hemostasis. However, when injected they may raise serum epinephrine levels and increase the potential for adverse cardiovascular effects.9  In contrast, although topical epinephrine may be absorbed through mucous membranes and bleeding tissues, it causes local vasoconstriction, which decreases systemic absorption. Administration of topical epinephrine has been shown to result in elevation of serum concentrations 140 times less than injections of even dilute epinephrine preparations.10,11 

Complications associated with use of topical epinephrine are extremely rare, and changes in cardiovascular outcomes have not been shown to be statistically significant.10,12,13  To the best of our knowledge, no well-designed clinical trials have been conducted to assess the cardiovascular effects of topical epinephrine on gingival tissue in a pediatric population.

The purpose of this pilot study was to determine the cardiovascular safety and hemostatic efficacy of topical racemic epinephrine compared to placebo. The primary objective was to determine if topical racemic epinephrine had any effect on heart rate (HR) or blood pressure in children receiving dental care under general anesthesia (GA). We hypothesized that topical racemic epinephrine would be associated with no significant changes in HR, blood pressure, or mean arterial pressure (MAP) compared to placebo. The secondary objective was to determine if topical racemic epinephrine had any effect on hemostasis, as measured subjectively by the treating dentist. We hypothesized that topical racemic epinephrine would result in more rapid hemostasis than placebo.

METHODS

Study Participants

The study was approved by the University of Washington Institutional Review Board (IRB STUDY00006670). Potential subjects were recruited from a pool of patients scheduled to receive comprehensive dental care under GA at the study clinic. Patients and their parents/legal guardians were approached regarding study participation on the day of the initial dental consultation or were contacted by phone at least 2 days prior to their scheduled surgery. Consent was obtained the day of the surgery. Inclusion criteria included American Society of Anesthesiologists (ASA) physical status class I or II, English speaking, and the presence of caries requiring prefabricated crowns on both primary maxillary first molars (#B and #I).

Subjects were excluded if parents/legal guardians were unable to communicate with the study coordinator in English or if the patient was ASA III+; had a history of cardiac arrhythmia, cardiovascular disease, diabetes, or thyroid disease; and/or was prescribed antiarrhythmic, antihypertensive, or ionotropic medications. Subjects were also excluded if they required pulpotomy or pulpectomy treatment on the primary maxillary first molars.

Study Design and Procedures

This pilot study was a single-blinded, crossover/split-mouth randomized controlled trial design. Patients were recruited from June until November 2019, and 16 children who met inclusion criteria were approached for participation. Three patients were excluded: 1 patient did not meet the inclusion criteria after new radiographs were taken under GA and 2 caregivers declined to participate in the study on the day of surgery (Figure 1).

Figure 1.

Participant flow in single-blinded, split-mouth randomized controlled pilot study.

Figure 1.

Participant flow in single-blinded, split-mouth randomized controlled pilot study.

Randomization was performed in 2 stages. Using Stata SE version 14.2 (StataCorp), we first randomized the side for the control treatment. The right primary maxillary first molar was assigned the control treatment, which was maintained for all participants. Treatment sequence (ie, control or experimental treatment performed first) was then randomly assigned to each participant. The randomization list was placed into a password-protected Excel file prior to the start of the study and then used to assign the order participants received treatment interventions.

This method permitted patients to serve as their own control and ensured randomization of treatment sequencing. Parents were not present during the procedure, and patients were under GA and therefore blinded to the treatment. However, the treating dentist and study personnel were not blinded.

Data Collection

On the day of surgery, after ensuring adherence to NPO guidelines and updating patient weight and medical history, participants were transferred to the operating room, where they underwent mask induction with sevoflurane 8%, nitrous oxide 50–70%, and oxygen. Standard ASA monitors consisting of a 5-lead electrocardiogram, noninvasive automated blood pressure cuff on the ankle or upper arm, pulse oximeter, capnography, and temperature probe were applied after induction and maintained throughout the course of the procedure. Vital signs were measured and recorded continuously.

Immediately after peripheral intravenous access was established, propofol 1–2 mg/kg, dexamethasone 4–6 mg, and ketorolac 0.5 mg/kg were administered intravenously, followed by direct laryngoscopy and nasotracheal intubation. Anesthesia was maintained with continuous infusions of propofol 50–100 mcg/kg/min and remifentanil 0.05–0.1 mcg/kg/min, plus nitrous oxide 30–70% and oxygen.

After successful intubation, necessary radiographs were taken and a throat pack was placed, followed by a dental cleaning, examination, and treatment planning. Next, the primary maxillary first molar was prepared for a zirconia crown based on the randomized treatment sequencing (control vs experimental). To reduce stimulation, no dental isolation (ie, rubber dam) was used during any of the study procedures. After preparation, an appropriately sized zirconia crown was fit, and baseline HR and blood pressure were recorded. Next, 2 cotton pellets were stretched and applied circumferentially to the gingival tissue of the prepared tooth. Pellets for the control treatment were prepared by soaking the pellets in 0.9% sodium chloride (physiological saline). Pellets for the experimental treatment (HemeRx, Racellet #3, Sprig Oral Health Technologies, Inc) were prepared as indicated by the manufacturer. They contain ∼0.55 mg (0.42–0.68 mg) of racemic epinephrine hydrochloride per pellet and are advertised for use in pediatric restorative procedures. Pellets were maintained in position for 1 minute with gauze pressure as directed per the manufacturer's recommendation. After pellet removal, any residual coagulum was removed using suction or moistened gauze.

Cardiovascular outcomes including the patient's HR, systolic blood pressure (SBP), diastolic blood pressure (DBP), and MAP were recorded immediately before pellet placement (baseline) and repeated at 1, 2, 3, 4, and 5 minutes after removal.

The adequacy of hemostasis was determined subjectively by the operating dentist/principal investigator (T.M.N.) as “adequate” or “inadequate” at baseline and again at 1, 2, 3, 4, and 5 minutes after removal. Adequate hemostasis was defined as cessation of hemorrhage from the gingival tissues. Inadequate hemostasis was defined as continued hemorrhage from the gingival tissues, with blood contamination of the prepared tooth. If hemostasis was not achieved after 5 minutes, lidocaine 2% and epinephrine 1:100,000 (0.01 mg/mL) were administered by injection and direct pressure was applied with gauze prior to crown cementation. The study procedures (preparation, crown sizing, pellet application, and hemostasis evaluation) were immediately repeated for the contralateral side using the alternative study treatment.

Following study procedures, the remainder of each patient's dental care was completed, including application of rubber dam isolation, sealants, composite restorations, pulp therapy, stainless-steel crowns, and extractions. All patients were discharged on the same day after adequate recovery and observation time.

Data Analysis

Data were analyzed using Stata SE version 14.2. Descriptive statistics are reported as mean and standard deviation values. Paired t test and 1-way repeated analysis of variance (ANOVA) were used for comparison of cardiovascular outcomes within the control and the experimental groups. A paired t test with equal variance was used for comparing the change (%) in cardiovascular measurements from baseline to 5 minutes for the control and experimental treatments as well as time to adequate hemostasis. The McNemar exact test was used to compare whether a patient reached adequate hemostasis following the control or experimental treatments. A P value <.05 was utilized to determine statistical significance.

RESULTS

Demographic Characteristics

The final sample included a total of 13 patients (54% males/46% females), ranging from 2.7 to 7.8 years old, with mean age of 4.7 ± 1.6 years and weight of 18.2 ± 4.2 kg. All patients were ASA class I (n = 10) or II (n = 3).

Cardiovascular Outcomes

All baseline cardiovascular measurements (HR, SBP, DBP, and MAP) recorded prior to pellet application were generally comparable between the control and experimental treatments (Tables 1 and 2).

Table 1.

Cardiovascular Measurements Following Saline Treatment (Control)*

Cardiovascular Measurements Following Saline Treatment (Control)*
Cardiovascular Measurements Following Saline Treatment (Control)*
Table 2.

Cardiovascular Measurements Following Topical Epinephrine Treatment (Experimental)*

Cardiovascular Measurements Following Topical Epinephrine Treatment (Experimental)*
Cardiovascular Measurements Following Topical Epinephrine Treatment (Experimental)*

Comparison of cardiovascular data from baseline to 5 minutes following the control treatment (saline) revealed significant decreases in HR (91.5 vs 88.5 bpm; P = .04) and SBP (89.0 vs 86.1 mm Hg; P = .047). However, no significant change was noted in DBP (37.8 vs 35.8 mm Hg; P = .13) or MAP (54.8 vs 53.2 mm Hg; P = .28). The 1-way ANOVA failed to demonstrate any significant differences in cardiovascular data over time from baseline through all time periods (Table 1).

Comparison of cardiovascular data from baseline to 5 minutes following the experimental treatment (topical racemic epinephrine) revealed significant decreases for all measurements, including HR (93.4 vs 89.6 bpm; P = .01), SBP (91.4 vs 88.8 mm Hg; P = .02), DBP (41.3 vs 36.1 mm Hg; P = .01) and MAP (58.2 vs 53.2 mm Hg; P = .01). The 1-way ANOVA failed to demonstrate any significant differences in cardiovascular data over time from baseline through all time periods (Table 2).

A comparison of the percentage change (baseline to 5 minutes postintervention) in cardiovascular data between the groups revealed significantly larger decreases in mean DBP (−11.1% vs −3.9%; P < .01) and MAP (−8.1% vs −2.1%; P < .01) following topical epinephrine treatment compared with saline control treatment (Table 3).

Table 3. 

Comparing Change (%) In Cardiovascular Parameters From Baseline to 5 Minutes Following the Saline (Control) or Topical Epinephrine (Experimental) Treatments

Comparing Change (%) In Cardiovascular Parameters From Baseline to 5 Minutes Following the Saline (Control) or Topical Epinephrine (Experimental) Treatments
Comparing Change (%) In Cardiovascular Parameters From Baseline to 5 Minutes Following the Saline (Control) or Topical Epinephrine (Experimental) Treatments

Neither treatment produced any clinically meaningful changes in cardiovascular measurements from baseline to 5 minutes. No subjects experienced any cardiac arrythmias or other adverse outcomes during the study procedures or following discharge home.

Hemostasis Outcomes

Figure 2A and B demonstrates examples of adequate versus inadequate hemostasis. All 13 teeth receiving the experimental treatment reached adequate hemostasis after an average of 2.2 minutes. Meanwhile, only 5 of the 13 teeth receiving the control treatment reached adequate hemostasis after an average of 4.2 minutes during the 5-minute observation period, which was a significant difference (P < .01). When comparing the 5 patients who reached hemostasis for both the experimental and control treatments, there was still a significant difference in time to achieve adequate hemostasis (1.6 vs 4.2 minutes; P = .01).

Figure 2.

Representative images of hemostasis and intraoral preop and postop views in study participants. (A) Adequate hemostasis of tooth #I; (B) inadequate hemostasis of tooth #I; (C) intraoral preop occlusal view; and (D) intraoral postop occlusal view with zirconia crowns on #B and #I.

Figure 2.

Representative images of hemostasis and intraoral preop and postop views in study participants. (A) Adequate hemostasis of tooth #I; (B) inadequate hemostasis of tooth #I; (C) intraoral preop occlusal view; and (D) intraoral postop occlusal view with zirconia crowns on #B and #I.

Eight of the 13 control teeth did not reach adequate hemostasis within 5 minutes. In these cases, hemostasis was successfully achieved following the local anesthetic injection and application of direct pressure with gauze prior to crown cementation. Figure 2C and D demonstrates preoperative and postoperative intraoral photos of the maxillary arch of a representative patient for this study.

DISCUSSION

The purpose of this pilot study was to determine the cardiovascular safety and hemostatic efficacy of topical racemic epinephrine pellets used to obtain hemostasis in pediatric patients following zirconia crown preparation. The results demonstrated a small, statistically significant but clinically irrelevant decrease in all cardiovascular parameters after topical application of the racemic epinephrine pellets. In addition, there was significantly less time needed to achieve adequate hemostasis using topical racemic epinephrine pellets. This finding indicates that compared to a saline placebo, topical application of 2 concentrated epinephrine pellets does not pose a substantial cardiovascular risk in healthy children treated under controlled GA conditions and is an effective method to achieve hemostasis.

In this pilot study, we observed a decrease in the cardiovascular parameters over the course of each 5-minute observation period following both control and experimental treatments (Table 3). This was likely because patients were maintained on a continuous infusion of general anesthetic medications during the 5-minute observation period and no clinically stimulating procedures were performed after recording baseline vital signs.

A statistically significant difference between DPB and MAP data for the control and experimental treatments was noted when comparing the mean percentage change between baseline and the 5-minute time point. However, these decreases were not clinically significant and did not lead to any changes in clinical management. When applied directly to gingival tissues, the cardiovascular effects of topical epinephrine pellets as used in this study appear to be negligible. However, there was a significant difference in hemostasis between tissues treated with epinephrine pellets and those treated with saline pellets. Hemostasis was achieved within ∼2.2 minutes in all cases when epinephrine pellets were used. In contrast, fewer than half of the sites treated with saline achieved hemostasis within 5 minutes. This finding is both statistically and clinically relevant.

Prefabricated zirconia crowns are still a relatively recent addition to the repertoire of restorative pediatric dentistry. Although scientific evidence regarding their performance is limited, recent findings indicate that prefabricated zirconia crowns are not only aesthetically superior but may also provide a more durable alternative to stainless steel crowns for primary molars,14  reduce plaque accumulation, and improve gingival health.6  However, zirconia crowns have reduced mechanical retention compared to stainless steel crowns or bonded composite, which may result in loss of the restoration. The preparation for zirconia crowns is also more aggressive, which may result in necrosis and pulp-related failures.15  Although no restorative treatment is perfect, prefabricated zirconia crowns are a durable and esthetic treatment option for primary teeth. As the market for esthetic pediatric restorative options grows, the need for safe and effective methods for obtaining rapid hemostasis may also increase. The results of this pilot study indicate that topical epinephrine pellets appear to be a safe and effective option for this purpose, at least at the dosages used in this study (0.42–0.68 mg/pellet). However, larger studies should be completed before generalizing these conclusions.

These findings are similar to those of prior published research investigating use of topical epinephrine to achieve rapid hemostasis. Studies by Korkmaz et al and Gunaratne et al showed no significant hypertensive episodes and no hemodynamic parameter changes associated with placement of up to 10 mg (1 mg/mL, 1:1000) of topical epinephrine on vascular sinus tissue.16,17  However, other studies have reported topical epinephrine inducing significant hemodynamic changes in small subsets of patients, particularly those with preexisting cardiovascular diseases.18,19 

A literature review investigating the effects of topical racemic epinephrine showed its effectiveness in decreasing intraoperative bleeding. Degerliyurt et al showed effective use of cotton sponges soaked with 24 to 48 mcg of racemic epinephrine (1:100,000) for sinus surgery without concerns for safety.10,20,21  Vickers et al studied the cardiovascular effects of topical epinephrine pellets containing ∼0.55 mg of racemic epinephrine per pellet (0.42–0.68 mg/pellet) and 20% ferric sulfate in endodontic surgery. They determined neither agent had any statistically significant cardiovascular effects, but the epinephrine-soaked pellets produced subjectively better hemostasis.12 

STRENGTHS AND LIMITATIONS

The single-blinded split-mouth/crossover study design was a major strength of this pilot study, enabling the performance of within-patient comparisons rather than strictly between-patient comparisons. Thus, the error variance of the experiment was reduced, thereby obtaining a more powerful statistical test with a small number of patients. However, the single-blinded study design contributed to potential performance bias for the study personnel and detection bias for outcome assessment, as the treating dentist could distinguish between treatments because of the difference in color of the epinephrine and saline pellets as well as the placement of the pellets (tooth #B or #I).

Another major strength of this pilot study was the use of GA, administered via continuous intravenous infusions and inhaled anesthetic agents. GA eliminated potential changes in vital signs from patient agitation, ensured ideal behavioral compliance, and enabled detection of minute alterations in vital signs. In addition to the aforementioned advantages, GA permitted completion of all restorative procedures without use of injected local anesthesia. In a clinical scenario with patients not undergoing GA, local anesthetic containing epinephrine would be used. This could potentially decrease time to achieve adequate hemostasis with both the control and experimental interventions. At the same time, the infusions used to maintain GA during the procedure may have introduced a degree of variability in cardiovascular function and therefore impacted clinical bleeding. All patients were kept at a level of anesthesia that prevented patient movement and facilitated spontaneous ventilation at 10–20 breaths/min. Therefore, although the depth of anesthesia likely varied amongst participants, the impact was minimized by maintaining a constant infusion rate and having each participant serve as their own control. Additionally, although ketorolac can negatively affect platelet function, it was administered to all patients prior to restorative treatment. It remains unclear if a single dose of ketorolac has an immediate, clinically significant effect on gingival bleeding.

The small sample size of this study limits the generalizability of the results and may have led to potential sampling bias. The investigators recognize this as a potential cause for concern and suggest that the results of this study should be interpreted cautiously. The methods used in this study may provide a starting point for future larger clinical trials.

CONCLUSION

In this randomized controlled pilot study, patients treated with topical racemic epinephrine pellets showed a statistically significant decrease in DBP and MAP compared with saline pellets 5 minutes after placement. However, the decreases in cardiovascular vital signs following application of saline or racemic epinephrine pellets were all clinically irrelevant. Adequate hemostasis was achieved more rapidly after using racemic epinephrine pellets than with saline pellets.

ACKNOWLEDGMENTS

We thank patients, their families, and staff at Dental Surgery Center for their participation and support.

REFERENCES

1. 
Centers for Disease Control and Prevention
.
Hygiene-related disease; dental caries (tooth decay)
.
2016
.
Available at: https://www.cdc.gov/. Accessed November 23, 2018.
2. 
American Academy of Pediatric Dentistry
.
Pediatric restorative dentistry. The Reference Manual of Pediatric Dentistry 2019–2020
.
Pediatr Dent
.
2019
:
340
352
.
3. 
Waggoner
WF.
Restoring primary anterior teeth: updated for 2014
.
Pediatr Dent
.
2015
;
37
:
163
170
.
4. 
Abdulhadi
B,
Abdullah
M,
Alaki
S,
Alamoudi
N,
Attar
M.
Clinical evaluation between zirconia crowns and stainless steel crowns in primary molars teeth
.
J Pediatr Dent
.
2017
;
5
:
21
27
.
5. 
Walia
T,
Salami
AA,
Bashiri
R,
Hamoodi
OM,
Rashid
F.
A randomized controlled trial of three aesthetic full-coronal restorations in primary maxillary teeth
.
Eur J Paediatr Dent
.
2014
;
15
:
113
118
.
6. 
Taran
PK,
Kaya
MS. A
Comparison of periodontal health in primary molars restored with prefabricated stainless steel and zirconia crowns
.
Pediatr Dent
.
2018
;
40
:
334
339
.
7. 
Phillips-Reed
LD,
Austin
PN,
Rodriguez
RE.
Pediatric tonsillectomy and ketorolac
.
J Perianesth Nurs
.
2016
;
31
:
485
494
.
8. 
Gobble
RM,
Hoang
HL,
Kachniarz
B,
Orgill
DP.
Ketorolac does not increase perioperative bleeding: a meta-analysis of randomized controlled trials
.
Plast Reconstr Surg
.
2014
;
133
:
741
755
.
9. 
Jastak
JT,
Yagiela
JA.
Vasoconstrictors and local anesthesia: a review and rationale for use
.
J Am Dent Assoc
.
1983
;
107
:
623
630
.
10. 
Degerliyurt
K,
Denizci
S.
Does the topical use of epinephrine for sinus floor augmentation affect systemic hemodynamics?
Implant Dent
.
2013
;
22
:
289
294
.
11. 
Kumar
M,
Chawla
R,
Goyal
M.
Topical anesthesia
.
J Anaesthesiol Clin Pharmacol
.
2015
;
31
:
450
456
.
12. 
Vickers
FJ,
Baumgartner
JC,
Marshall
G.
Hemostatic efficacy and cardiovascular effects of agents used during endodontic surgery
.
J Endod
.
2002
;
28
:
322
323
.
13. 
van Hasselt
CA,
Low
JM,
Waldron
J,
Gibb
AG,
Oh
TE.
Plasma catecholamine levels following topical application versus infiltration of adrenaline for nasal surgery
.
Anaesth Intensive Care
.
1992
;
20
:
332
336
.
14. 
Kist
S,
Stawarczyk
B,
Kollmuss
M.
Fracture load and chewing simulation of zirconia and stainless-steel crowns for primary molars
.
Eur J Oral Sci
.
2019
;
127
:
369
375
.
15. 
Seminario
AL,
Garcia
M,
Spiekerman
C,
Rajanbabu
P,
Donly
KJ,
Harbert
P.
Survival of zirconia crowns in primary maxillary incisors at 12-, 24- and 36-month follow-up
.
Pediatr Dent
.
2019
;
41
:
385
390
.
16. 
Gunaratne
DA,
Barham
HP,
Christensen
JM,
Bhatia
DD,
Stamm
AC,
Harvey
RJ.
Topical concentrated epinephrine (1:1000) does not cause acute cardiovascular changes during endoscopic sinus surgery
.
Int Forum Allergy Rhinol
.
2016
;
6
:
135
139
.
17. 
Korkmaz
H,
Yao
WC,
Korkmaz
M,
Bleier
BS.
Safety and efficacy of concentrated topical epinephrine use in endoscopic endonasal surgery
.
Int Forum Allergy Rhinol
.
2015
;
5
:
1118
1123
.
18. 
Orlandi
RR,
Warrier
S,
Sato
S,
Han
JK.
Concentrated topical epinephrine is safe in endoscopic sinus surgery
.
Am J Rhinol Allergy
.
2010
;
24
:
140
142
.
19. 
Yim
MT,
Ahmed
OG,
Takashima
M.
Evaluating real-time effects of topical 1:1000 epinephrine in endoscopic sinus and skull-base surgery on hemodynamic parameters through intraoperative arterial line monitoring
.
Int Forum Allergy Rhinol
.
2017
;
7
:
1065
1069
.
20. 
Higgins
TS,
Hwang
PH,
Kingdom
TT,
Orlandi
RR,
Stammberger
H,
Han
JK.
Systematic review of topical vasoconstrictors in endoscopic sinus surgery
.
Laryngoscope
.
2011
;
121
:
422
432
.
21. 
Cohen-Kerem
R,
Brown
S,
Villasenor
LV,
Witterick
I.
Epinephrine/lidocaine injection vs. saline during endoscopic sinus surgery
.
Laryngoscope
.
2008
;
118
:
1275
1281
.