Secondary Logo

Journal Logo

Original Article

Prehospital intravenous line placement assessment in the French emergency system

a prospective study

Minville, V.*; Pianezza, A.*; Asehnoune, K.; Cabardis, S.*; Smail, N.*

Author Information
European Journal of Anaesthesiology: July 2006 - Volume 23 - Issue 7 - p 594-597
doi: 10.1017/S0265021506000202
  • Free



The out-of-hospital peripheral intravenous (i.v.) line placement and specifically the time consumed performing this procedure is still matter of debate in the Advanced Life Support (ALS) in the US-American Emergency Medical System (EMS) [1]. Effectively, the loss of time in severe trauma patients can lead to a fatal outcome [1]. In the US-American system, the procedure for i.v. placement must be applied by paramedics, who often provide over treatment of patients by placement of an i.v. line when a saline lock or no i.v. would suffice, resulting in unnecessary costs for EMS systems [2]. Moreover, the incidence of unused out-of-hospital i.v. placements is high in the US-American EMS, that is to say i.v. infusions are frequently started and not used in the emergency department [3]. However, even if out-of-hospital pain management (where an i.v. is needed) is an important EMS function, few patients transported by US-American EMS with musculoskeletal injuries receive out-of-hospital analgesia [4,5]. Several studies were undertaken in paramedics [2,6–9], but even if the difference between British/US-American and French/ German EMS continue to fuel the debate [10], no evaluation of peripheral i.v. access (PIA) in the French EMS exists concerning the success rate and the time to have a PIA. Moreover, in out-of-hospital cases in France, where a lot of catheters are used each day, French physicians are present in ALS units that could lead to a better indication of PIA.

The purpose of this prospective study was to quantify the precise time required to achieve PIA in the French ALS EMS, and to assess the use and feasibility of the PIA placement in this system.


We performed a prospective and descriptive (observational) study during a period of 3 months, day and night, in ALS patients (trauma and medical) to assess the PIA. After approval by the institutional review board, the study was conducted at a University Hospital in urban and suburban ALS (population approximately 1 000 000). Five vehicles were involved in the study. In the French EMS a physician is present in the ambulance, accompanied by a nurse and a driver. In the university hospital, there are students on board. All the patients who had PIA attempted by ALS unit were consecutively included in this study. There were no exclusion criteria. The ALS units involved in this study were not randomly sampled because all units were included.

The following items were recorded: i.v. line attempts, i.v. line time procedure interval, i.v. line success rate, number and size of catheter used, intubation, blood sampling, fluid and drugs used.

Different practitioners were allowed to perform the i.v. line (doctor, nurse, medical students). Doctors and nurses were experienced in the practice of i.v. placement and medical students had theoretical and practical tuition in the operating room before the study. The i.v. line procedure interval was defined as the time elapsed between the placement of a tourniquet until the i.v. line was taped and secured (using a stopwatch). An i.v. line was defined as successful when a catheter had been taped in place and found to be patent. An i.v. line was defined as unsuccessful when the attempts at i.v. cannulation did not lead to blood reflux. An i.v. line attempt was defined as the use of a catheter to perform one cannulation and counted as one attempts. Complications (e.g. infiltration, bleeding) were also collected.

Statistical analyses were performed using the Statview software (version 5.0, California, USA). Data are presented as mean ± SD or percent. A χ2-test or a t-test was performed when appropriate. P < 0.05 was considered statistically significant.


The mean age of the patients was 58 ± 20 yr. There were 64 trauma victims (17%), 139 cardiology (36%) including 90 chest pain (23%), 49 cardiac arrest (10%), 31 toxicology (8%) and 105 patients with various medical problems (29%). Four hundred and twenty-seven PIA were placed, i.e. 10% of patients had several PIA.

Three hundred and eighty-eight patients were included of 669 patients involved by ALS (58% had a PIA). The main outcome of the study was PIA success rate. Seventy-six percent of patients had a patent i.v. after one attempt, and 98% had a patent i.v. after a second attempt. The total success rate was 99.7%, only one patient was impossible to cannulate after four attempts and was transported without PIA. The success rate and length of PIA starting time were not associated with the patient's categorization as trauma or non-trauma (medical). Sixty-three percent of the PIA were placed by nurses, 29% by doctors and 8% by students. Statistically significant difference was found between doctors and students with regard to the time to place the i.v. as well as for success rate (P < 0.05). A statistically significant difference was found between nurses and students in regard to the time to place the i.v. as well as for the success rate (P < 0.05). No significant difference was found between doctors and nurses. The mean time for i.v. line procedure was 4.4 ± 2.8 min with a range of 1–8.58% of patients underwent blood sampling that statistically increased time of procedure (4.3 ± 3 min vs. 4.6 ± 2.5 min, P < 0.05).

Sixty-three percent of PIA were sized 18-G, 31% were sized 20-G, 6% were sized 16-G and <1% were sized 22-G and 14-G. Fifty-one percent received isotonic saline infusion, 44% received glucose solution infusion and 5% received fluid infusion alone for hypotension. Fluid infusion was 9% of hydroxyethylstarch in 13% of cases, Ringers lactate in 3% of cases, gelofusine in 0.7% of cases and hypertonic saline solution in 0.3% of cases. Drug was infused in 71% of cases. Eighteen percent of patients were transported without a doctor on board after resolving the medical or trauma problem (mainly analgesia). More than three drugs (from three to six) were infused into a single PIA. All PIA were performed on scene. No complication was found during the study.


We prospectively studied attempts at i.v. line placement in 388 patients by ALS in the French out-of-hospital system; and found that the average time required for a successful PIA was 4.4 min and the global success rate was 99.7%. The mean procedure interval range is from 1.3 [8] to 12 min [11] in the literature, the time recorded in our study was on the lower end of this range. It is difficult to compare the different mean time procedure interval because of the different criterion of i.v. line placement interval definition. In this study the i.v. line placement interval definition is very strict because we measured all the procedure from the tourniquet placement to the taped and secured i.v. line (i.e. time to obtain functional i.v. line access).

The overall success rate for i.v. line placement (99.7%) is similar to that found in the literature (98.3%, 91% and 89%) [3,8,9]. Only one study described an overall success rate of 100% [7]. Only one failure to place an i.v. line was found in a 58-yr-old male with chest pain and normal electrocardiogram (ECG), and a past medical history of dialysis for chronic renal failure (poor venous access). The patient was taken to the emergency without i.v. following assessment by a doctor. Paramedics would have probably have persevered because the protocol that says that patients with chest pain require an i.v.

Blood sampling could be considered as a time-consuming procedure. However, it should be kept in mind that the availability of the nurses might be increased in these overloaded departments, so the turnover of patients might be improved. Some studies have reported that i.v. access plus blood sampling can be accomplished in less than 2.5 min [7,12,13]. Moreover, while statistically significant in our study, a difference of 0.3 min (4.3 ± 3 min vs. 4.6 ± 2.5 min) is not clinically relevant.

No complications were noticed during this study related to the PIA. Nevertheless radial nerve injury or infectious complications have been related to radial PIA [14–17]. Finally, when the patients were hospitalized, PIA placed during this study was withdrawn within 24 h after the placement avoiding possible infection of the puncture site. Effectively, even if in France there are recommendations concerning the PIA in regard to aseptic procedure [18], these recommendations are not well followed [19]. Lapostolle and colleagues found that doctors and students are farther from these recommendations than nurses [19]. They can be explained by the strict education of the nurses concerning aseptic procedures.

In our study analysis, two groups appeared. First the patient group with a drug infusion, second the patient group without a drug infusion. The first group, i.e. therapeutic PIA, represented two-thirds of patients. In this group a single i.v. line was placed in 88% of the patients, while greater than three drugs were infused in 21% of the cases. Twenty-five percent of these patients were incubated, ventilated and sedated, so another PIA should have been placed avoiding bolus related hypotension, if another drug infusion was required [20]. In the same way, a catecholamine bolus should have been avoided in the patients transported to the hospital with a catecholamine continuous infusion.

The second group, i.e. security PIA, represented one-third of patients. No significant difference was found for mean time procedure interval and overall success rate between this group and the drug infusion group. Physicians had decided to place these PIA to prevent loss of time for the patient in case of an urgent situation (e.g. epilepsy). In fact, while physicians are on board the medical problem is often resolved completely or in the critical part at the arrival at the hospital. On the contrary, for trauma patient surgery is frequently needed, prehospital physicians perform the resuscitation or provide correct analgesia during transportation [21]. Out-of-hospital physicians probably provide better analgesia than paramedics [4,5], however, it needs to be studied in a double-blind fashion. In the US and many other paramedics-based systems, the decision is based on protocol. One of the criticisms of this type of practice is that it likely results in too many precautionary or security i.v. lines being placed. In the French EMS, we have no protocols regarding out-of-hospital i.v. placement. Using the discretion of the physician on the ALS units results in a lower percentage of inappropriate precautionary lines. Thus a gold standard for inappropriateness could be found but require a further study.

Our study has several limitations. We only had conducted this observational study in a single university hospital. Perhaps we should have excluded students from the study because a student is not always on board in all the French EMS. However, it can be informative because of the lack of study concerning this subject, the only study being that of Lapostolle and colleagues [19]. Moreover, the practice in a prehospital setting is the same all over the country and the facility is typical of the entire country, so this study should be representative of the regular behaviour of the intensivist in France with i.v. line placement in out-of-hospital care. Another limitation is that in some cases, anaesthesiologists staff out-of-hospital units. Although we evaluated our institutional strategy of PIA, we did not compare the success rates of individual anaesthesiologists or emergency medicine physicians. As we did not intend to perform that comparison, we did not record data that would allow us to determine this potential difference. A final limitation is that it takes place only in the French EMS. Thus, the results cannot be extended to EMS in other countries where paramedics are present in the ambulances. A comparative study comparing other systems could be very interesting.

In conclusion, we prospectively studied attempts at i.v. line placement in 388 patients by French out-of-hospital units and found that the average time required for a successful PIA was 4.4 min and that out-of-hospital i.v. line placement in the French EMS is safe with a good overall success rate (99.7%).


1. Sampalis JS, Tamim H, Denis R et al. Ineffectiveness of on-site intravenous lines: is prehospital time the culpit? J Trauma 1997; 43: 608–615.
2. Gausche M, Tadeo RE, Zane MC, Lewis RJ. Out-of-hospital intravenous access: unnecessary procedures and excessive cost. Acad Emerg Med 1998; 5: 878–882.
3. Pace SA, Fuller FP, Dahlgren TJ. Paramedic decisions with placement of out-of-hospital intravenous lines. Am J Emerg Med 1999; 17: 544–547.
4. White LJ, Cooper JD, Chambers RM, Gradisek RE. Prehospital use of analgesia for suspected extremity fractures. Prehosp Emerg Care 2000; 4: 205–208.
5. McEachin CC, McDermott JT, Swor R. Few emergency medical services patients with lower-extremity fractures receive prehospital analgesia. Prehosp Emerg Care 2002; 6: 406–410.
6. Jones SE, Nesper TP, Alcouloumre E. Prehospital intravenous line placement: a prospective study. Ann Emerg Med 1989; 18: 244–246.
7. Slovis CM, Herr EW, Londorf D et al. Success rates for initiation of intravenous therapy en route by prehospital care providers. Am J Emerg Med 1990; 8: 305–307.
8. Pons PT, Moore EE, Cusick JM et al. Prehospital venous access in an urban paramedic system – a prospective on-scene analysis. J Trauma 1988; 28: 1460–1463.
9. Spaite DW, Valenzuela TD, Criss EA et al. A prospective in-field comparison of intravenous line placement by urban and non urban emergency medical services personnel. Ann Emerg Med 1994; 24: 209–214.
10. Dick WF. Anglo-American vs. Franco-German emergency medical services system. Prehospital Disaster Med 2003; 18: 29–35.
11. Border JR, Lewis FR, Aprahamian C et al. Panel: prehospital trauma care-stabilize or scoop and run. J Trauma 1983; 23: 708–711.
12. McSwain GR, Garrison WB, Artz C. Evaluation of resuscitation from cardiopulmonary arrest by paramedics. Ann Emerg Med 1980; 9: 341–345.
13. Feldman R. i.v. line placement: a time study for prehospital providers. J Emerg Med 1986; 11: 43–45.
14. Thrush DN, Belsole R. Radial nerve injury after routine peripheral vein cannulation. J Clin Anesth 1995; 7: 160–162.
15. Mimoz O, Rayeh F, Debaene B. Catheter-related infection in intensive care. Physiopathology, diagnosis, treatment and prevention. Ann Fr Anesth Réanim 2001; 20: 520–536.
16. Levine R, Spaite DW, Valenzuela TD et al. Comparison of clinically significant infection rates among prehospital-versus in-hospital-initiated i.v. lines. Ann Emerg Med 1995; 25: 502–506.
17. Norwood S, Ruby A, Civetta J, Cortes V. Catheter-related infections and associated septicaemia. Chest 1991; 99: 968–975.
18. Agence nationale d'accréditation et d'évaluation en santé (Anaes). Evaluation des pratiques professionnelles dans les établissements de santé. Evaluation de la qualité de la pose et de la surveillance des cathéters veineux courts. Juin 1998.
19. Lapostolle F, Garrigue B, Monmarteau V et al. Peripheral venous placement: evaluation of guidelines respect in prehospital setting. Ann Fr Anesth Reanim 2005; 24: 31–35.
20. Ricard-Hibon A, Chollet C, Belpomme V et al. Epidemiology of adverse effects of prehospital sedation analgesia. Am J Emerg Med 2003; 21: 461–466.
21. Lopez S, Gros T, Bernard N et al. Fascia iliaca compartment block for femoral bone fractures in prehospital care. Reg Anesth Pain Med 2003; 28: 203–207.


© 2006 European Society of Anaesthesiology