Central Venous Port Implantation

Free consultation

Ask how your port will be placed and where the tip will sit

Send your diagnosis and treatment plan including which drugs are scheduled, any previous chest or neck surgery or radiotherapy, any previous central line and what happened to it, a note of any clotting disorder or blood thinner, and any recent chest imaging with the image files. A surgeon reviews the file and tells you which vein is suitable, whether ultrasound and tip confirmation are part of the plan here, and what the schedule for flushing will be after you go home. No fee, no obligation, and a coordinator replies in your own language, usually within the same working day.

What the first hour decides

Two things are being decided while you are on the table. The first is whether the needle finds the vein cleanly, which determines whether you leave with a collapsed lung or a hematoma in your neck. The second is where the end of the catheter comes to rest inside your chest, which influences whether the device works reliably for the next two years. Neither is visible to you afterward, and both are recorded in your operative note.

Pooling 35 studies and 5,108 patients, guiding the needle with ultrasound instead of feeling for landmarks reduced total complications with a risk ratio of 0.29, accidental arterial puncture with a risk ratio of 0.28 across 22 trials and 4,388 patients, and hematoma with a risk ratio of 0.27. Success at the first attempt rose with a risk ratio of 1.57, and the average number of attempts fell by more than one.

Companion reviewers looking at the vein under the collarbone instead of the one in the neck found the same direction of effect, with arterial puncture at a risk ratio of 0.21 and hematoma at 0.26 across three trials and 498 patients. Both reviews cover central lines of every kind instead of implanted ports specifically, and the physical act being studied is identical. What neither review can tell you is how much of that benefit survives in the hands of an operator who never used landmarks in the first place, because the trials compared techniques and not operators. What can be said is that the effect is large, that it has been found repeatedly in trials of every size, and that no serious argument has been made against it in twenty years, which puts it in a small category of surgical questions that are genuinely settled.

Every decision, and its cost

Eight decisions get made about your port, four of them before you wake up and four of them repeatedly for as long as you keep it. This table is the shape of the whole page.

Wide table. Swipe or drag it sideways on a narrow screen, because it scrolls instead of shrinking.

What a port is

Which vein

What the intensive care trial found

Most of the quoted evidence on venous site comes from a trial that randomized 3,471 catheters in 3,027 intensive care patients between the vein under the collarbone, the vein in the neck and the vein in the groin. Complication rates ran at 1.5, 3.6 and 4.6 per 1,000 catheter days respectively. Compared with the collarbone route, the groin carried a hazard ratio of 3.5 with a confidence interval of 1.5 to 7.8, and the neck carried a hazard ratio of 2.1 which reached significance at a p value of 0.04. The collarbone route paid for its advantage in collapsed lungs, at thirteen requiring a chest drain against four in the neck group. That trial studied short-term catheters in critically ill patients, and a port that stays in for two years in someone having chemotherapy every three weeks is a different situation. Dwell time, the reason the line is there and the state of the person it is in all differ, and any of the three could reverse the ranking of the sites, which is why the trial belongs on this page as background instead of as an instruction.

What the port trial found

One randomized trial has asked the question in port patients directly, allocating 403 people with cancer between three approaches, which were the neck vein by landmark, the collarbone vein under ultrasound and a surgical cut-down onto the cephalic vein in the groove of the shoulder. Median follow-up ran to 356.5 days. Early complications were absent in both the neck and collarbone arms and occurred in 1.5 percent of the cut-down arm. Thrombosis occurred in fifteen, eight and eleven patients respectively, a spread that looks meaningful and did not reach significance at a p value of 0.272. The trial's own conclusion was that neither the site nor the method changed early or late complication rates, and the only clear finding was that the ultrasound-guided collarbone approach failed less often, at a p value of 0.001. Its title promises a best choice of insertion site and its data do not deliver one, which is worth noticing whenever the paper is cited in support of a particular vein.

How to read those two together

With 403 patients and almost no early complications, that trial is not powered to detect a difference between sites, so the honest reading is that no site advantage was demonstrated rather than that all sites are equal. What did emerge is that how the vein is entered mattered more than which vein was entered, which is the thread running through the next three sections. If your surgeon has a settled preference and can explain it, that preference is not the thing to argue about.

Cut-down or puncture

Getting the catheter into a vein can be done two ways. One opens a small incision and threads it into a vein under direct vision. The other pushes a needle through the skin into a deeper vein. A German trial of 1,205 patients settled part of this and left part of it open.

• Open cut-down caused fewer collapsed lungs. Among 1,159 analyzed patients the odds ratio for pneumothorax or blood in the chest cavity was 0.27, with a confidence interval of 0.09 to 0.88 and a p value of 0.029. This was a superiority trial and superiority was established, which is worth stating because the trial is filed in one major database under a label suggesting it tested for equivalence.
• Everything else was the same. The trial reported no significant differences in primary success, tolerability, overall illness after surgery, radiation dose or death within thirty days. The closed approach was quicker. So the whole case for cut-down rests on one complication, and that complication is the one nobody wants.
• Cut-down also fails more often. A pooled analysis of six randomized trials covering 1,831 patients found an odds ratio of 0.308 for pneumothorax favoring cut-down and an odds ratio of 2.364 against it for failure of the intended technique, with a confidence interval of 1.051 to 5.315 that barely excludes one. Those two results have to be quoted together or not at all.
• The comparison was not entirely fair. In the German trial the puncture arm was subclavian puncture without routine ultrasound guidance. Given what the ultrasound evidence shows, that is not the modern alternative, and the authors bounded their own conclusion to comparable cohorts. The sensible synthesis is that cut-down is a good first choice and ultrasound-guided puncture is the right second line when it fails.
• These two papers are not independent. The 1,159 patients of the randomized trial make up roughly two thirds of the 1,831 in the pooled analysis, and the two share most of their authors. Counting them as two separate confirmations of the same finding would overstate the evidence, and that is a mistake worth avoiding when a single result is doing the work.

Why ultrasound matters

The catheter that breaks

A catheter that passes between the collarbone and the first rib gets squeezed every time you move your shoulder. Over months that repeated compression can wear through the tubing, and the loose end travels with the bloodstream into the heart or the pulmonary artery. This is called pinch-off, and it belongs specifically to the route beneath the collarbone.

In a series of 1,320 subclavian port placements, nine catheters ruptured and embolized, which is 0.68 percent. Only two of the nine had a visible pinch-off sign on x-rays reviewed afterward. Eight of the nine presented with painful swelling around the port during an infusion, and none had any symptom from the fragment itself. Every fragment was retrieved through a vein in the groin, and nobody died.

Something in that paragraph matters more than the headline number, and it comes in two parts. The first is that the radiological warning sign is unreliable, described by the authors themselves as far from constant, so a normal x-ray does not exclude the problem. The second is what the patients actually noticed, which was pain and swelling at the port during an infusion rather than anything dramatic. That is the symptom to report immediately, and it is the same symptom you would get from a much more ordinary problem, which is why it is worth reporting instead of diagnosing yourself.

Published estimates put the overall incidence of fracture and embolization at somewhere between 0.5 and 1 percent, drawn from the wider literature instead of from their own cohort. It is rare, it is fixable, and it is one of the arguments for choosing a route that does not pass through that narrow space at all. Cut-down onto a vein at the shoulder and puncture of the internal jugular vein in the neck both avoid the costoclavicular gap entirely, and that is a point in their favor which has nothing to do with the trials described further up this page.

What large series report

Five published series covering more than seven thousand devices between them give a reasonable picture of what actually happens. The numbers vary because the populations and the definitions vary, and the shape is consistent.

Another wide one. Swipe or drag it sideways on a narrow screen, because it scrolls instead of shrinking.

Questions to take with you

The first six belong in the conversation before implantation. The last four belong in the conversation about living with the device afterward.

1. Is ultrasound used routinely for the puncture, or only after a first attempt has failed?
2. Which vein and which technique are planned for me, and why that one?
3. Tell me how the position of the catheter tip will be confirmed before I leave the operating room.
4. Will I get a copy of the operative note recording the vein, the technique and the tip position?
5. When after implantation will the port first be used?
6. Give me your own rate of collapsed lung and of failed placement for this operation.
7. What solution will be used to lock the port, and at what interval when I am not on treatment?
8. Who is allowed to access it, and what should I say if someone reaches for an ordinary needle?
9. Which symptoms should make me call the same day instead of waiting for the next appointment?
10. When would you plan to take it out, and what happens if I would rather keep it longer?

Where the tip should sit

The target, and why it is a target

Catheters are meant to end at the junction where the superior vena cava meets the right atrium. Too high and the tip lies against the vein wall in slower-moving blood, where drugs are less diluted and clot is more likely to form. Too low and it is inside the heart. The window is not enormous, and it is why the position gets checked before the operation finishes instead of the next morning. Radiologists and surgeons describe the target slightly differently and the practical result is the same, which is that the tip should be in the lower third of the superior vena cava or just inside the atrium, and that a tip sitting well above that is a finding worth acting on instead of accepting.

The catheter moves when you do

Investigators compared the tip position in 62 chest ports on films taken lying down and standing up, and found the tip migrated about 20 millimeters upward, a difference with a p value of 0.003. That single finding explains a great deal about why ports that looked perfectly positioned on the day sometimes stop drawing blood weeks later, and it is the reason experienced operators aim slightly lower than the textbook target when the patient is lying flat. Your position check happens while you are flat on a table, and you then spend the rest of the device's life upright, so a tip that looked correct during the procedure sits about two centimeters higher for most of its working life. The same study looked at whether higher tips malfunctioned more and reported 36 percent against 8 percent, and that comparison rested on four events against two and did not reach statistical significance at a p value of 0.057. It is a reasonable hypothesis and it is not a demonstrated fact, and the authors said as much.

Confirming the position without x-rays

An electrical method exists in which the catheter itself is used as an electrode and the shape of the heart's electrical trace tells the operator when the tip has reached the right place. In a study of 231 adults having ports placed, the proportion of ideal tip positions did not differ between this method and x-ray screening, at a p value of 0.733, though both methods were used in the same patients so this is an agreement study instead of a comparison of strategies. A larger multicenter study of the technique agreed with imaging in 95.8 percent of cases, and that study was conducted in children and mostly in devices other than ports, so it does not transfer directly to an adult having a chest port. What both suggest is that the electrical method is a reasonable alternative where fluoroscopy is unavailable, and no more than that. Neither study measured whether patients did better afterward.

Infection

Infection is the commonest reason a port has to come out. It ran at 4.7 percent in a German series of 1,008 devices, 1.44 percent in a Chinese series of 2,996 breast cancer patients, and 0.21 per 1,000 catheter days in a Taiwanese series of 1,449 ports. Those are not contradictory figures so much as three different ways of counting over three different follow-up periods, and a rate per catheter day is the only one that lets you compare across studies honestly. A percentage tells you what share of patients had a problem without saying over how long, and since a port that stays in for two years has roughly four times the exposure of one that stays in for six months, two identical percentages can describe very different devices.

Two interventions have been tested properly and they point in opposite directions. Skin antisepsis works. In a trial of 2,349 intensive care patients, cleaning the skin with chlorhexidine in alcohol rather than iodine in alcohol reduced catheter-related infection from 1.77 to 0.28 per 1,000 catheter days, a hazard ratio of 0.15 with a confidence interval of 0.05 to 0.41. Scrubbing the skin first added nothing. Severe skin reactions were more common with chlorhexidine, at 27 patients against 7, and led to stopping it in two, so it is worth mentioning a previous reaction before anyone reaches for the bottle. That trial studied short-term catheters in critically ill patients whose baseline infection rate is far above that of an oncology port, so the hazard ratio should not be read as a port figure, and the choice of antiseptic it supports applies every time your port is accessed.

Giving antibiotics before implantation does not work. Pooling five randomized trials and 360 adults, the risk ratio for infection was 0.72 with a confidence interval running from 0.33 to 1.58 and a p value of 0.41. If a unit routinely gives you an antibiotic before putting a port in, it is doing something that has been tested and found not to help. The same review did find a benefit from locking the device with an antibiotic solution rather than heparin, at a risk ratio of 0.47 and a number needed to treat of twelve where the baseline infection rate is around 15 percent, and its authors stated plainly that this evidence came from 468 participants who were mostly children and that they could not tell whether it applied to fully implanted ports at all. That caveat has to travel with the finding every time, because an antibiotic lock is not a small intervention and the population in which it was shown to work is not the population reading this page.

Clots

How often it happens

Clot forms either as a sleeve of fibrin around the catheter, which is common and often silent, or as a thrombosis of the vein itself, which swells the arm and the neck on that side. In the large series above, deep vein thrombosis of the arm accounted for 0.63 percent of patients in one, 1.9 percent in another and 3.6 percent in a third, and fibrin around the catheter was found in 1.84 percent. Cancer itself raises clotting risk, so separating the contribution of the device from the contribution of the disease is not straightforward.

Whether blood thinners prevent it

Thirteen randomized trials covering 3,420 people with cancer and central lines have tested this, and the answer splits by drug. Low molecular weight heparin reduced symptomatic catheter-related clot with a risk ratio of 0.43 and a confidence interval of 0.22 to 0.81, which works out at 38 fewer events per thousand people, and the reviewers graded that as moderate certainty. Warfarin did not, with a risk ratio of 0.61 whose confidence interval ran from 0.23 to 1.64 and therefore included no effect at all. Neither drug reduced deaths. Comparing the two directly, low molecular weight heparin caused more low platelet counts, with a risk ratio of 1.69 and a confidence interval of 1.20 to 2.39. Those thirteen trials span tunnelled catheters, ports and arm lines together and the review does not break the results out by device, so the numbers describe central lines in cancer patients as a category rather than ports specifically, which is a limitation the reviewers state and most secondary accounts drop.

Why nobody gives it routinely

Preventing 38 clots per thousand people sounds worthwhile until you notice that the same analysis found no reduction in deaths and that the burden falls on every patient while the benefit falls on a few. The reviewers put it as a balance the patient should weigh, which is unusually direct language for a systematic review. Guidelines do not recommend routine prevention, and none of this evidence covers the newer oral anticoagulants, which were not eligible for the review and have not been tested for this purpose.

Heparin or saline

Every port is flushed after use and locked with a solution that sits in the catheter until next time. Whether that solution needs to contain heparin has been argued for thirty years, and the argument is not over. Four studies are usually cited and they do not agree.

Wide again. Drag the table sideways on a narrow screen, because it scrolls instead of shrinking.

How often it needs flushing

The case for stretching the interval

If you have finished treatment and are keeping the port in case the disease returns, a flush every four weeks means twelve hospital visits a year for nothing else. A Korean study followed 154 colorectal cancer patients flushed at a mean interval of 98.4 days and found the port still working in 152 of them, with 25 of the 30 patients who did relapse able to use the existing device without a new operation. That is a genuinely useful result for anyone weighing a year of monthly appointments.

Why it is weaker than it looks

The Korean study began with 214 patients and excluded 60, of whom 54 were removed for violating the flushing interval. In other words, everyone in whom the three month schedule was not maintained was taken out of the count before the outcome was measured, so the 152 out of 154 is a result conditional on perfect adherence and cannot be read as what happens in ordinary practice. There was also no comparison group flushed more often, so the study cannot show that three months is as good as one. A separate randomized trial of 143 patients compared heparin every four months, heparin every two months and saline every two months, and found no blockages at all until month ten, when a single patient was withdrawn. An absence of events in a small trial is not a demonstration of equivalence, and the authors' own equivalence threshold was never assessed. Zero events is a comfortable finding and it is not a proof.

What to ask beforehand

What happens on the day

The rules week to week

Once the port is in, the things that determine whether it lasts are small, repeated and mostly outside your control. Knowing what they are lets you notice when they are not being done.

This table is wide too. Swipe it sideways on a narrow screen, because it scrolls instead of shrinking.

What to send us

Whether a port is straightforward in your case is largely answerable from documents, and the awkward cases are usually flagged by something in this list.

• Your diagnosis and the treatment schedule. Which drugs, how often, and for how long. A regimen given every three weeks for six months makes different demands on a device than continuous infusion pumps or long term parenteral feeding, and one large series found the combination of chemotherapy and feeding through the same port had a much lower survival than chemotherapy alone.
• Anything that has been done to your chest or neck before. Previous surgery, radiotherapy, a previous port or line on either side, and what happened to it. Scarred or previously thrombosed veins change the plan entirely and are the commonest reason a first choice is abandoned.
• Recent chest imaging, with the image files. A computed tomography scan done for staging usually shows the veins well enough to plan the route. Send the images and not only the report, because the radiologist here will want to look at the vessels themselves.
• Clotting history and current blood thinners. Any previous clot, any known clotting disorder, and the exact drug and dose instead of the category. Plans differ by drug and the timing around the procedure has to be arranged in advance.
• Which side you would rather it went. Handedness, a seatbelt, a bra strap, a rifle stock, a violin. This is the one decision on the whole page where your preference is the deciding evidence, and nobody will guess it.

When it comes out

How long they last

In a series of 324 ports followed over eleven years and accumulating 159,764 catheter days, median follow-up was 490.6 days and removal for a complication happened in 8.6 percent. In a French series of 815 patients followed for a year, 55 devices came out early, at a mean of 3.7 months after implantation, and neither series describes an object that fails often, and both describe one where most failures happen relatively early instead of at the end of a long working life.

Why they come out

Infection led the French list at 19 removals, followed by the device working its way out through the skin at 14, catheter migration at 6, clot at 5 and mechanical problems at 3. The German series of 1,008 devices reported 32 early complications leading to removal in 11 and 88 late complications leading to removal in 34, which is to say that most complications did not cost the patient the device. That is worth knowing before you assume a problem means another operation.

Keeping it after treatment ends

Roughly three in ten patients relapse after finishing treatment for some cancers, mostly within two years, and that is the case for keeping the port rather than removing it at the end of chemotherapy. In the Korean surveillance study, 25 of the 30 patients who did relapse used the existing port for their new treatment without needing another operation. Against that sits the maintenance burden, the small ongoing infection risk and the fact that a device sitting under your skin is a daily reminder. There is no evidence that settles this one, and it is a reasonable conversation to have instead of a decision to accept by default. What tips it for many people is the prospect of a second operation at the worst possible moment, since a relapse is not the point at which anyone wants to be arranging vascular access from scratch, and that argument carries more weight than any of the numbers on either side.

Coming to Istanbul

Central venous port FAQ